JP2024545146A - SYSTEMS, DEVICES AND METHODS FOR COMPENSATING FOR TEMPERATURE EFFECTS ON SENSORS - Patent application - Google Patents

Abstract

Various examples are directed to systems and methods for generating an estimated analyte value, such as a glucose value. The analyte sensor system may access a first sensor signal from an in vivo analyte sensor and a first temperature signal from an in vitro temperature sensor. The analyte sensor system may generate a first analyte sensor temperature based at least in part on the first temperature signal and generate a first estimated analyte value based at least in part on the first sensor signal and a first temperature compensated sensitivity. Further examples are directed to using a non-analyte background signal.

Description

CLAIM OF PRIORITY This application claims priority to U.S. Patent Application No. 17/545,711, filed December 8, 2021, which is incorporated by reference herein in its entirety.

The present development relates generally to medical devices such as analyte sensors, and more specifically, but without limitation, to systems, devices, and methods for compensating for the effects of temperature on analyte sensors.

Diabetes is a metabolic disease linked to the body's production or use of insulin. Insulin is a hormone that allows the body to use glucose for energy or store it as fat.

When a person eats a meal containing carbohydrates, the food is processed by the digestive system, producing glucose in the blood. Blood glucose can be used for energy or stored as fat. The body normally maintains blood glucose levels in a range that provides enough energy to support bodily functions and avoids problems that can result from glucose levels that are too high or too low. Regulation of blood glucose levels depends on the production and use of insulin, which regulates the movement of blood glucose into cells.

When the body does not produce enough insulin or is unable to effectively use the insulin that is present, blood glucose levels can rise above the normal range. A condition or state in which blood glucose levels are higher than normal is called "hyperglycemia." Chronic hyperglycemia can lead to many health problems, including cardiovascular disease, cataracts and other eye diseases, nerve damage (neuropathy), and kidney damage. Hyperglycemia can also lead to acute problems, such as diabetic ketoacidosis (a condition or state in which the body becomes too acidic due to the presence of glucose in the blood and ketone bodies, which are produced when the body can no longer use glucose). A condition or state in which blood glucose levels are lower than normal is called "hypoglycemia." Severe hypoglycemia can lead to acute crises that can result in seizures or death.

Diabetic patients may receive insulin to manage blood glucose levels. Insulin may be received, for example, through a manual injection with a needle. Wearable insulin pumps are also available. Diet and exercise also affect blood glucose levels. Glucose sensors provide estimated glucose values and can be used as guidance by patients and caregivers.

Diabetes is sometimes referred to as "Type 1" and "Type 2." Type 1 diabetes patients are typically able to use insulin when it is present, but due to problems with the insulin-producing beta cells of the pancreas, the body is unable to produce enough insulin. Type 2 diabetes patients are able to produce some insulin, but due to reduced sensitivity to insulin, the patient is "insulin resistant." As a result, even though insulin is present in the body, the insulin is not used well by the patient's body and does not effectively regulate blood sugar levels.

This Background is provided to introduce a brief context for the Summary and Detailed Description that follow. This Background is not intended to aid in determining the scope of the claimed subject matter, nor should it be deemed to limit the claimed subject matter to implementations that solve any or all of the disadvantages or problems presented above.

This specification discusses, among other things, systems, devices, and methods for determining subcutaneous temperature or compensating for the effects of temperature on an analyte sensor, such as a glucose sensor.

One example (e.g., "Example 1") of the subject matter (e.g., system) may include determining a temperature-compensated glucose concentration level by receiving a temperature signal indicative of a temperature parameter of an external component, receiving a glucose signal indicative of an in vivo glucose concentration level, and determining the compensated glucose concentration level based on the glucose signal, the temperature signal, and a delay parameter.

In Example 2, the subject matter of Example 1 can be optionally configured such that the temperature parameter is a temperature, a temperature change, or a temperature offset.

In Example 3, the subject matter of either Example 1 or 2 may optionally be configured such that the temperature parameter is detected at a first time and the glucose concentration level is detected at a second time after the first time, where the delay parameter may be configured to include a delay period between the first time and the second time that accounts for a delay between a first temperature change in the external component and a second temperature change proximate the glucose sensor.

In Example 4, the subject matter of any one or more of Examples 1-3 may optionally include adjusting the delay period based on a rate of temperature change.

In Example 5, the subject matter of any one or more of Examples 1-4 may optionally include adjusting the delay period based on the detected condition or state.

In Example 6, the subject matter of any one or more of Examples 1-5 may optionally be configured such that the detected condition or state includes a sudden change in temperature.

In Example 7, the subject matter of Examples 5 or 6 may optionally be configured such that the detected condition or state includes movement.

In Example 8, the subject matter of any one or more of Examples 1-7 may optionally be configured such that detecting the glucose signal includes receiving the glucose signal from a wearable glucose sensor.

In Example 9, the subject matter of Example 8 may optionally be configured such that detecting the temperature signal includes measuring a temperature parameter of a component of the wearable glucose sensor.

In Example 10, the subject matter of Examples 8 or 9 may optionally be configured such that determining the compensated glucose concentration level includes executing instructions on a processor to receive a glucose signal and a temperature signal, and determining the compensated glucose concentration level using the glucose signal, the temperature signal, and the delay parameter.

In Example 11, the subject matter of any one or more of Examples 8-10 may optionally include storing a value corresponding to the temperature parameter in the memory circuit and retrieving the stored value from the memory circuit for use in determining the compensated glucose concentration level.

In Example 12, the subject matter of any one or more of Examples 1-11 may optionally include delivering a therapy based at least in part on the compensated glucose concentration level.

One example (e.g., "Example 13") of the subject matter (e.g., a system) may include a glucose sensor circuit configured to generate a glucose signal representative of a glucose concentration level, a temperature sensor circuit configured to generate a temperature signal indicative of a temperature parameter, and a processor configured to determine a compensated glucose concentration level based on the glucose signal, the temperature signal, and the delay parameter.

In Example 14, the subject matter of Example 13 can be configured such that the temperature parameter is a temperature, a temperature change, or a temperature offset.

In Example 15, the subject matter of Examples 13 or 14 may be configured such that the delay parameter includes a delay period that accounts for a delay between a first temperature change in the temperature sensor circuit and a second temperature change in the glucose sensor circuit.

In Example 16, the subject matter of Example 15 may be configured such that the processor adjusts the delay period based on a rate of change of temperature determined using the temperature parameter.

In example 17, the subject matter of examples 15 or 16 may be configured such that the processor adjusts the delay period based on the detected condition or determined state.

In Example 18, the subject matter of any one or any combination of Examples 13-17 may be configured such that the processor executes instructions to receive the glucose signal and the temperature signal and apply a delay parameter to determine a compensated glucose concentration level.

In Example 19, the subject matter of any one or any combination of Examples 13-19 may further include a memory circuit, and the memory circuit may be configured such that the system stores a value corresponding to the temperature parameter in the memory circuit and the processor subsequently retrieves the stored value from the memory for use in determining the compensated glucose concentration level.

In Example 20, the subject matter of any one or any combination of Examples 13-19 may be configured such that the glucose sensor circuit includes an electrode operably coupled to an electronic circuit configured to generate a glucose signal and a membrane covering at least a portion of the electrode, the membrane including an enzyme configured to catalyze a reaction between glucose and oxygen from a biological fluid in contact with the membrane in vivo.

An example (Example 21) of a subject (e.g., a system, device, or method) for determining a temperature-compensated glucose concentration level may include receiving a glucose sensor signal, receiving a temperature parameter signal, receiving a third sensor signal, evaluating the temperature parameter signal using the third sensor signal to generate an evaluated temperature parameter signal, and determining a temperature-compensated glucose concentration level based on the evaluated temperature parameter signal and the glucose sensor signal.

In Example 22, the subject matter of Example 21 may be configured such that receiving the third sensor signal includes receiving a heart rate signal.

In Example 23, the subject matter of Examples 21 or 22 may be configured such that receiving the third signal includes receiving a blood pressure signal.

In Example 24, the subject matter of any one or any combination of Examples 21-23 may be configured such that receiving the third signal includes receiving an activity signal.

In Example 25, the subject matter of any one or any combination of Examples 21 to 24 may be configured such that receiving the third sensor signal includes receiving a position signal.

In Example 26, the subject matter of any one or any combination of Examples 21-25 may be configured such that evaluating the temperature parameter signal includes determining the presence at a location of known temperature characteristics.

In Example 27, the subject matter of any one or any combination of Examples 21-26 may be configured such that the method includes determining the presence at a location having known ambient temperature characteristics.

In Example 28, the subject matter of any one or any combination of Examples 21-27 may be configured such that the method includes determining the presence of a location having a water-logged environment.

In Example 29, the subject matter of Example 28 can be configured such that the submerged environment is a pool or a beach.

In Example 30, the subject matter of any one or any combination of Examples 21-29 may be configured such that receiving the third sensor signal includes receiving temperature information from an ambient temperature sensor.

In Example 31, the subject matter of any one or any combination of Examples 21 to 30 may be configured such that receiving the third sensor signal includes receiving information from a wearable device.

In Example 32, the subject matter of Example 31 may be configured such that receiving the third sensor signal includes receiving information from a wristwatch.

In example 33, the subject matter of any one or any combination of examples 21-32 may be configured such that receiving the third sensor signal includes receiving temperature information from a physiological temperature sensor. In some examples, the subject matter may include a watch or other wearable device that includes a temperature sensor.

In Example 34, the subject matter of any one or any combination of Examples 21-33 may be configured such that receiving the temperature parameter signal includes receiving a signal indicative of a temperature, a temperature change, or a temperature offset.

In Example 35, the subject matter of any one or any combination of Examples 21-34 may be configured such that receiving the third signal includes receiving an accelerometer signal.

In Example 36, the subject matter of any one or any combination of Examples 21-35 may further include detecting motion using a third signal.

In Example 37, the subject matter of any one or any combination of Examples 21-36 may be configured such that evaluating the temperature parameter signal includes determining that a change in the temperature parameter signal is consistent with an exercise session.

In Example 38, the subject matter of any one or any combination of Examples 21-37 may be configured such that evaluating the temperature parameter signal includes determining that the temperature parameter signal is consistent with an occurrence of an increase in body temperature due to exercise.

In Example 39, the subject matter of any one or any combination of Examples 21-38 may be configured such that determining the temperature-compensated glucose concentration level includes applying the temperature parameter signal to an exercise model.

In Example 40, the subject matter of any one or any combination of Examples 21-39 may be configured such that the method includes applying the motion model when motion is detected and a change in the temperature parameter signal indicates a reduction in temperature (e.g., which may indicate motion in a cool or convection-cooled environment).

In Example 41, the subject matter of any one or any combination of Examples 21-40 may be configured such that the third signal includes a heart rate signal, a respiration signal, a blood pressure signal, or an activity signal, and movement is detected from an increase in the heart rate signal, the respiration signal, the blood pressure signal, or the activity signal.

A glucose sensor of the subject matter (e.g., system, device, or method) of one example ("Example 42") configured to generate a first signal representative of a glucose concentration in a host, the glucose sensor including a temperature sensor configured to generate a second signal representative of a temperature, and a processor that evaluates the second signal based on the third signal and generates a temperature-compensated glucose concentration level based at least in part on the evaluation of the first signal and the second signal.

In example 43, the subject matter of example 42 may be configured such that the processor evaluates the second signal by corroborating the detected temperature or temperature change using the third signal.

In example 44, the subject matter of examples 42 or 43 may be configured such that the processor determines a condition or state based on the third signal and validates the detected temperature or temperature change based on the condition or state.

In Example 45, the subject matter of any one or any combination of Examples 42-44 may be configured such that the condition or state is a location, an ambient environment, an activity condition or state, or a physiological condition.

In Example 46, the subject matter of any one or any combination of Examples 42-45 may be configured such that the processor suspends temperature compensation based at least in part on the third signal.

In Example 47, the subject matter of any one or any combination of Examples 42-46 may be configured such that the processor is configured to detect movement based at least in part on the third signal.

In example 48, the subject matter of example 47 may be configured such that in response to detecting movement, the processor may be configured to suspend temperature compensation despite a drop in temperature indicated by the second signal, and the processor may be configured to avoid inaccurate temperature compensation when the host is exercising in a low temperature (e.g., cold outdoor or convection cooled) environment.

In Example 49, the subject matter of any one or any combination of Examples 42-48 may be configured such that the processor specifies a temperature compensation model based at least in part on the third signal.

In Example 50, the subject matter of any one or any combination of Examples 42-49 may further include a third sensor, the third sensor generating a third signal.

In Example 51, the subject matter of any one or any combination of Examples 42-50 may be such that the third signal includes location information, and the processor is configured to evaluate the second signal based at least in part on the location information.

In Example 52, the subject matter of any one or any combination of Examples 42-51 can be such that the third signal includes activity information, and the processor is configured to evaluate the second signal based at least in part on the activity information.

In Example 53, the subject matter of any one or any combination of Examples 42-52 may be configured such that the temperature compensated glucose sensor system includes a wearable continuous glucose monitor including a glucose sensor and a temperature sensor.

In Example 54, the subject matter of any one or any combination of Examples 42-53 may be configured such that the temperature compensated glucose sensor system includes an activity sensor and the third signal includes activity information from the activity sensor.

In example 55, the subject matter of any one or any combination of examples 42-54 may be configured such that the third signal includes the host's heart rate, respiratory rate, or blood pressure.

In example 56, the subject matter of any one or any combination of examples 42-55 may be configured such that the processor detects movement based on changes in heart rate, respiration rate, or blood pressure.

In Example 57, the subject matter of any one or any combination of Examples 42-56 may be configured such that the processor is configured to confirm the elevated body temperature indicated by the second signal based at least in part on the detection of movement.

In example 58, the subject matter of any one or any combination of examples 42-56 may be configured such that the processor is configured to decrease, taper, limit, or suspend temperature compensation in response to detecting motion.

In Example 59, the subject matter of any one or any combination of Examples 42-58 may be configured such that the third signal includes a signal from an optical sensor configured to detect a blood parameter of the host.

In Example 60, the subject matter of Example 59 may further include an optical sensor, the optical sensor including a light source and a photodetector configured to detect blood flow rate or red blood cell count in a region of the host beneath the optical sensor.

One example ("Example 61") of the subject matter (e.g., a system, device, or method) may include temperature compensating the continuous glucose sensor by determining a pattern from the temperature data, receiving a glucose signal from the continuous glucose sensor, the glucose signal indicative of a glucose concentration level, and determining a temperature-compensated glucose concentration level based at least in part on the sensor glucose signal and the pattern.

In example 62, the subject matter of example 61 may be configured such that determining the pattern includes determining a pattern of temperature fluctuations, and the method includes compensating the glucose concentration level according to the pattern.

In Example 63, the subject matter of Examples 61 or 62 may further include receiving a temperature parameter, comparing the temperature parameter to the pattern, and determining a temperature-compensated glucose concentration level based at least in part on the comparison.

In example 64, the subject matter of example 63 can be configured such that the pattern includes a temperature pattern that correlates with a physiological cycle.

In Example 65, the subject matter of Examples 63 or 64 may be configured such that the method includes determining whether the temperature parameter is highly reliable based on the comparison to the pattern, and temperature compensating the glucose concentration level using the temperature parameter when the temperature parameter is determined to be highly reliable.

In Example 66, the subject matter of any one or any combination of Examples 63-65 may be configured such that the method includes determining a degree of compensation based at least in part on a comparison of the temperature parameter to the pattern. For example, the degree of compensation may be based on a defined range or confidence interval.

In Example 67, the subject matter of any one or any combination of Examples 61-66 may be configured such that determining the pattern includes determining a condition or state, and determining the temperature-compensated glucose concentration level is based at least in part on the determined condition or state.

In example 68, the subject matter of example 67 may be configured such that determining the condition or state includes applying the temperature parameter to the state model.

In example 69, the subject matter of examples 67 or 68 may be configured such that determining the condition or state includes applying one or more of glucose concentration level, carbohydrate sensitivity, time, activity, heart rate, respiratory rate, posture, insulin delivery amount, meal time, or meal size to a state model.

In Example 70, the subject matter of any one or any combination of Examples 67-69 may be configured such that determining the condition or state includes determining the motion condition or state, and the method includes adjusting the temperature compensation-based model based on the motion condition or state.

One example ("Example 71") of the subject matter (e.g., a system, device, or method) may include a glucose sensor circuit configured to generate a glucose signal representative of a glucose concentration level, a temperature sensor circuit configured to generate a temperature signal indicative of a temperature parameter, and a processor that receives the glucose signal and the temperature signal and determines a temperature-compensated glucose concentration level based at least in part on a pattern determined from the glucose signal and the temperature signal.

In example 72, the subject matter of example 71 may be configured such that the processor determines a temperature parameter based on the temperature signal, compares the temperature parameter to the pattern, and determines a temperature-compensated glucose concentration level based at least in part on the comparison.

In Example 73, the subject matter of Examples 71 or 72 may be configured such that the processor determines whether the temperature parameter is highly reliable based on the comparison to the pattern, and if the temperature parameter is determined to be highly reliable, the temperature parameter is used to temperature compensate the glucose concentration level.

In example 74, the subject matter of examples 72 or 73 may be configured such that the processor determines the extent of compensation based at least in part on a comparison of the temperature parameter to the pattern.

In example 75, the subject matter of any one or any combination of examples 71-74 may be such that the pattern includes a state model, and the processor is configured to determine a temperature-compensated glucose concentration level based at least in part on applying the temperature parameter to the state model.

In example 76, the subject matter of example 75 may be configured such that the processor determines a temperature-compensated glucose concentration level by additionally applying one or more of glucose concentration level, carbohydrate sensitivity, time, activity, heart rate, respiration rate, posture, insulin delivery rate, meal time, or meal size to the state model.

In example 77, the subject matter of examples 75 or 76 may be configured such that the processor determines motion conditions or states and adjusts the temperature compensation model based at least in part on the motion conditions or states.

In Example 78, the subject matter of any one or any combination of Examples 71-77 may further include a memory circuit including executable instructions for determining a pattern from the temperature signal and determining a temperature-compensated glucose concentration level based on the pattern, and the processor is configured to retrieve the instructions from the memory and execute the instructions.

In Example 79, the subject matter of any one or any combination of Examples 71-78 may be configured such that the processor receives information about the pattern from a remote system via the communications circuitry.

In example 80, the subject matter of example 79 may be configured such that the remote system receives temperature parameter information based on the temperature signal and determines a pattern from the temperature parameter information.

One example ("Example 81") of the subject matter (e.g., a method, system, or device) may include determining a first value from a first signal indicative of a temperature parameter of a component of a continuous glucose sensor system, receiving a glucose sensor signal indicative of a glucose concentration level, comparing the first value to a reference value, and determining a temperature compensated glucose level based on the glucose sensor signal and a comparison of the first signal to the reference value.

In example 82, the subject matter of example 81 may be configured such that the method includes determining a temperature difference from a reference condition or state based on a variation of the first value from a reference value without calibrating the temperature to the reference value.

In example 83, the subject matter of examples 81 or 82 may further include determining a reference value from the first signal.

In example 84, the subject matter of example 83 can be configured such that the continuous glucose sensor system includes a glucose sensor insertable into a host, and the baseline value is determined during a specified period of time following insertion of the glucose sensor into the host.

In example 85, the subject matter of examples 83 or 84 may be configured such that the continuous glucose sensor system includes a glucose sensor insertable into the host, and the baseline value is determined during a specified period of time following activation of the glucose sensor.

In Example 86, the subject matter of any one or any combination of Examples 83-85 may be configured such that the reference value is determined during the manufacturing process.

In Example 87, the subject matter of any one or any combination of Examples 83-86 can be configured such that the method includes determining a reference value during a first period of time and determining the first value during a second period of time, the second period of time occurring after the first period of time. The reference value can be, for example, a long-term average and the first value can be a short-term average.

In Example 88, the subject matter of Example 87 may further include updating the reference value based on one or more temperature signal values obtained in a third time period after the second time period.

In Example 89, the subject matter of any one or any combination of Examples 83 to 88 may be configured such that determining the reference value includes determining an average of a plurality of sample values obtained from the first signal.

In Example 90, the subject matter of any one or any combination of Examples 81-89 may be configured such that the temperature-compensated glucose level is determined based at least in part on a temperature-dependent sensitivity value that varies based on the deviation of the first value from a reference value.

One example ("Example 91") of the subject matter (e.g., a system, device, or method) may include a glucose sensor circuit configured to generate a glucose signal representative of a glucose concentration level, a temperature sensor circuit configured to generate a first signal indicative of a temperature parameter, and a processor, and may determine a temperature-compensated glucose level based on the glucose signal and a deviation of the first signal from a reference value.

In example 92, the subject matter of example 91 may be configured such that the processor determines the deviation of the first signal from a reference value without determining a temperature corresponding to the reference value.

In example 93, the subject matter of examples 91 or 92 may be configured such that the processor determines the reference value based on the first signal.

In Example 94, the subject matter of Example 93 may be configured such that the processor determines the reference value based on a plurality of sample values obtained from the first signal during the first time period.

In Example 95, the subject matter of Examples 93 or 94 may be configured such that the processor determines the reference value based on a plurality of sample values obtained from the first signal during a specified period of time after activation or insertion of the glucose sensor.

In Example 96, the subject matter of any one or any combination of Examples 93 to 95 may be configured such that the processor iteratively updates the reference value.

In Example 97, the subject matter of any one or any combination of Examples 91-96 may be configured such that the processor determines the reference value as an average of a plurality of sample values obtained from the first signal during a specified time period.

In Example 98, the subject matter of any one or any combination of Examples 91-97 may be configured such that the processor determines a temperature-compensated glucose level based on the glucose signal and a temperature-dependent sensitivity value that varies based on deviation from a reference value.

In Example 99, the subject matter of any one or any combination of Examples 91-98 may be configured such that the processor is configured to determine the temperature compensated glucose concentration level based on the model, and the model may be configured such that the glucose sensor value determined from the glucose signal and the sample value based on the first signal are applied to the model.

In Example 100, the subject matter of any one or any combination of Examples 91-100 may further include a memory circuit and executable instructions stored on the memory circuit for determining a temperature compensated glucose concentration level based on the glucose signal and a deviation of the first signal from a reference value.

One example (Example 101) of the subject matter (e.g., a method, system, or device) may include receiving a glucose signal indicative of a glucose concentration level, receiving a temperature signal indicative of a temperature parameter, detecting a condition or state, and determining a temperature-compensated glucose concentration level based at least in part on the glucose signal, the temperature signal, and the detected condition or state.

In example 102, the subject matter of example 101 can be configured such that the condition or state includes a high rate of change of the glucose signal, and temperature compensation is reduced or suspended during the period in which the glucose signal is experiencing the high rate of change.

In example 103, the subject matter of examples 101 or 102 may be configured such that the condition or state includes a sudden change in the temperature signal.

In example 104, the subject matter of example 103 may be configured such that temperature compensation is reduced or suspended in response to detecting a sudden change in temperature.

In example 105, the subject matter of examples 103 or 104 may be configured such that determining the temperature-compensated glucose concentration level includes using a previous temperature signal value in place of the temperature signal value associated with the sudden change in temperature.

In Example 106, the subject matter of any one or any combination of Examples 103-105 may be configured such that determining the temperature-compensated glucose concentration level includes determining an extrapolated temperature signal value based on a previous temperature signal value and using the extrapolated temperature signal value in place of the temperature signal value associated with the sudden change in temperature.

In example 107, the subject matter of example 106 can be configured such that in response to detecting a sudden change in temperature, a delay model is invoked, the delay model specifying a delay period for use in determining a temperature-compensated glucose level.

In example 108, the subject matter of any one or any combination of examples 101-107 may be configured such that the condition or state is the presence of radiant heat on the continuous glucose monitoring system.

In example 109, the subject matter of any one or any combination of examples 101-108 may be configured such that the condition or state is heat, and temperature compensation is reduced or suspended in response to detection of heat.

In example 110, the subject matter of example 109 can be configured such that the condition or state includes movement.

In example 111, the subject matter of example 110 may be configured such that the method includes decreasing, tapering, limiting, or pausing temperature compensation if motion is detected.

In example 112, the subject matter of any one or any combination of examples 101-111 may be configured such that the method includes using a linear model to determine the temperature-compensated glucose concentration level.

In example 113, the subject matter of example 112 may further include receiving a blood glucose calibration value, and the temperature compensation gain and offset are updated when the blood glucose calibration value is received.

In example 114, the subject matter of any one or any combination of examples 101-113 may be configured such that the method includes using a time series model to determine the temperature compensated glucose concentration level.

In example 115, the subject matter of any one or any combination of examples 101-114 may be configured such that the method includes using a partial differential equation to determine the temperature compensated glucose concentration level.

In example 116, the subject matter of any one or any combination of examples 101-115 may be configured such that the method includes using a probabilistic model to determine the temperature-compensated glucose concentration level.

In Example 117, the subject matter of any one or any combination of Examples 101-116 may be configured such that the method includes using the state model to determine the temperature-compensated glucose concentration level.

In Example 118, the subject matter of any one or any combination of Examples 101-117 may be configured such that the condition or state includes a body mass index (BMI) value.

In Example 119, the subject matter of any one or any combination of Examples 101-118 may be configured such that the method includes determining a long-term average using the temperature signal, and the temperature-compensated glucose concentration level is determined using the long-term average.

In example 120, the subject matter of any one or any combination of examples 101-119 may be configured such that a glucose signal indicative of a condition or state is received from a continuous glucose sensor, the condition or state being a pressure on the continuous glucose sensor.

In example 121, the subject matter of example 120 can be configured such that compression is detected based at least in part on a rapid drop in the glucose signal.

In example 122, the subject matter of examples 120 or 121 may be configured such that the condition or state is pressure during sleep.

In Example 123, the subject matter of any one or any combination of Examples 101-122 may be configured such that the condition or state is sleep.

In Example 124, in the subject matter of Example 123, sleep is detected using one or more of temperature, posture, activity, and heart rate, and the method includes applying a designated glucose alert trigger based on the detected sleep.

In Example 125, the subject matter of any one or any combination of Examples 101-124 may further include delivering insulin therapy, the therapy being determined based at least in part on the temperature-compensated glucose level.

One example ("Example 126") of the subject matter (e.g., a system, device, or method) may include a glucose sensor circuit configured to generate a glucose signal representative of a glucose concentration level, a temperature sensor circuit configured to generate a temperature signal indicative of a temperature parameter, and a processor configured to determine a compensated glucose concentration level based on the glucose signal, the temperature signal, and the detected condition or state.

In example 127, the subject matter of example 126 may be configured such that the condition or state includes a high rate of change of the glucose signal, and the processor may be configured to reduce, suspend, taper, or limit temperature compensation during the period of the high rate of change of the glucose signal.

In example 128, the subject matter of examples 126 or 127 may be configured such that the condition or state includes a sudden change in the temperature signal, and the processor may be configured to reduce, suspend, taper, or limit temperature compensation in response to detecting the sudden change in temperature.

In Example 129, the subject matter of any one or any combination of Examples 126-128 may be configured such that the condition or state includes motion, and when motion is detected, the processor may be configured to decrease, taper, limit, or suspend temperature compensation.

In Example 130, the subject matter of any one or any combination of Examples 126-129 may further include a second temperature sensor circuit configured to detect radiant heat on the continuous glucose monitoring system, and the detected condition or state includes radiant heat detected by the second temperature sensor circuit.

One example ("Example 131") of the subject matter (e.g., a device, system, or method) may include an elongated portion having a distal end configured for in vivo insertion into a host and a proximal end configured to operably couple to a circuit, and a temperature sensor at the proximal end of the elongated portion.

In example 132, the subject matter of example 131 can be configured such that the temperature sensor includes a thermistor.

In example 133, the subject matter of examples 131 or 132 may be configured such that the temperature sensor includes a temperature variable resistance coating.

In Example 134, the subject matter of any one or any combination of Examples 131 to 133 may be configured such that the temperature sensor includes a thermocouple.

In Example 135, the subject matter of Example 134 can be configured such that the elongated portion includes a first wire extending from the proximal end to the distal end, and the thermocouple includes the first wire and a second wire joined to the first wire to form the thermocouple.

In Example 136, the subject matter of Example 135 can be configured such that the first wire is tantalum or a tantalum alloy and the second wire is platinum or a platinum alloy.

In example 137, the subject matter of examples 135 or 136 may further include a transmitter coupled to the glucose sensor, where a first electrical contact on the transmitter is coupled to the first wire and a second electrical contact on the transmitter is coupled to the second wire.

One example ("Example 138") of the subject matter (e.g., a method, system, or device) may include receiving a calibration value of a temperature signal, receiving a temperature signal indicative of a temperature parameter from a temperature sensor, receiving a glucose signal indicative of a glucose concentration level from a continuous glucose sensor, and determining a temperature-compensated glucose concentration level based at least in part on the glucose signal, the temperature signal, and the calibration value.

In example 139, the subject matter of example 138 may be configured such that receiving a calibration value for the temperature signal includes obtaining the calibration during a manufacturing step having a known temperature.

In example 140, the subject matter of examples 138 or 139 may be configured such that receiving a calibration value for the temperature signal includes obtaining a temperature during a specified period of time following insertion of the continuous glucose sensor into the host.

One example ("Example 141") of the subject matter (e.g., a method, system, or device) may include receiving a temperature signal indicative of a temperature of a component of a continuous glucose sensor on a host, and determining an anatomical location of the continuous glucose sensor on the host based at least in part on the received temperature signal.

In example 142, the subject matter of example 141 can be configured such that the anatomical location is determined based at least in part on the sensed temperature.

In example 143, the subject matter of examples 141 or 142 may be configured such that the anatomical location is determined based at least in part on the variability of the temperature signal.

One example ("Example 144") of the subject matter (e.g., a method, system, or device) may include receiving a temperature signal indicative of a temperature parameter from a temperature sensor on a continuous glucose monitor, and determining from the temperature signal that the continuous glucose monitor has been restarted.

In Example 145, the subject matter of Example 144 may be configured such that determining from the temperature signal that the continuous glucose monitor has been restarted includes comparing a first temperature signal value before sensor initiation to a second temperature signal value after sensor initiation, and declaring the continuous glucose monitor to have been restarted when the comparison meets a similarity condition.

In Example 146, the subject matter of Examples 144 or 145 may be configured such that the similar condition is a temperature range.

One example ("Example 147") of the subject matter (e.g., a system, device, or method) may include a glucose sensor circuit configured to generate a glucose signal representative of a glucose concentration level, a temperature sensor circuit configured to generate a temperature signal indicative of a temperature parameter, a thermal deflector configured to deflect heat from the temperature sensor circuit, and a processor configured to determine a compensated glucose concentration level based at least in part on the glucose signal and the temperature signal.

One example ("Example 148") of the subject matter (e.g., a system, device, or method) may include a glucose sensor circuit configured to generate a glucose signal representative of a glucose concentration level of a host, a first temperature sensor circuit configured to generate a first temperature signal indicative of a first temperature parameter proximate the host, a second temperature sensor circuit configured to generate a second temperature signal indicative of a second temperature parameter, and a processor configured to determine a compensated glucose concentration level based at least in part on the glucose signal, the first temperature signal, and the second temperature signal.

In Example 149, the subject matter of Example 148 can be configured such that the processor determines a compensated glucose concentration level based in part on a temperature gradient between the first temperature sensor circuit and the second temperature sensor circuit.

In example 150, the subject matter of examples 148 or 149 may be configured such that the processor determines a compensated glucose concentration level based in part on an estimate of the heat flux between the first temperature sensor circuit and the second temperature sensor circuit.

In Example 151, the subject matter of any one or any combination of Examples 148-150 may be configured such that the second temperature circuit is configured to generate a temperature signal indicative of the ambient temperature.

In Example 152, the subject matter of any one or any combination of Examples 148-151 may be configured such that the processor is configured to generate a temperature signal indicative of a temperature of a transmitter coupled to the glucose sensor circuit.

One example ("Example 153") of the subject matter (e.g., a method, device, or system) may include receiving a glucose signal from a glucose sensor representative of a glucose concentration level of a host, receiving a first temperature signal indicative of a first temperature parameter proximate the host or the glucose sensor, receiving a second temperature signal indicative of a second temperature parameter, and determining a compensated glucose concentration level based at least in part on the glucose signal, the first temperature signal, and the second temperature signal.

In example 154, the subject matter of example 153 can be configured such that the first temperature signal is received from a first temperature sensor coupled to the glucose sensor and the second temperature signal is received from a second temperature sensor coupled to the glucose sensor.

In example 155, the subject matter of example 154 can be configured such that the compensated glucose concentration level is determined based at least in part on a temperature gradient between the first temperature sensor and the second temperature sensor.

In example 156, the subject matter of examples 154 or 155 may be configured such that the compensated glucose concentration level is determined based at least in part on a heat flux between the first temperature sensor and the second temperature sensor.

In Example 157, the subject matter of any one or any combination of Examples 154-156 may further include detecting an increase in the first temperature signal and a decrease in the second temperature signal, and adjusting the temperature compensation model based on the detected increase and decrease.

In example 158, the subject matter of example 157 may be configured such that the method includes detecting motion (e.g., outdoor motion or convection cooling motion) based at least in part on the detected rises and falls, and adjusting or applying a temperature compensation model based on the detection of the motion.

In Example 159, the subject matter of any one or any combination of Examples 154-158 may further include determining, based at least in part on the second temperature signal, that the temperature change is due to radiant or ambient heat, and adjusting or applying a temperature compensation model based on the determination.

One example ("Example 160) of the subject matter (e.g., a method, system, or device) may determine a glucose concentration level by receiving a temperature sensor signal, receiving a glucose sensor signal, applying the temperature sensor signal and the glucose sensor signal to a model, and receiving an output from the model related to the glucose concentration level, where the model compensates for multiple temperature dependent effects on the glucose sensor signal.

In example 161, the subject matter of example 161 can be configured such that the output is a compensated glucose concentration level.

In example 162, the subject matter of example 161 may further include delivering a therapy based on the compensated glucose concentration value.

In example 163, the subject matter of example 161 may be configured such that the model compensates for two or more of the sensor sensitivity, the local glucose level, the compartment bias, and the non-enzymatic bias. In some examples, the model may compensate for three or more of the sensor sensitivity, the local glucose level, the compartment bias, and the non-enzymatic bias. In some examples, the model may consider additional temperature-dependent factors in addition to the sensor sensitivity, the local glucose level, the compartment bias, and the non-enzymatic bias.

One example (Example 164) of the subject matter (e.g., a method, system, or device) may include determining an estimated analyte value level by determining a first value indicative of the conductance of a sensor component, determining a second value indicative of the conductance of the sensor component, receiving a signal representative of the estimated analyte value of the host, and determining a compensated estimated analyte value level based at least in part on a comparison of the second value to the first value. The first value and the second value may be, for example, electrical conductance or electrical resistance or electrical impedance.

In example 165, the subject matter of example 164 may be configured such that determining the first value includes determining an average conductance.

In Example 166, the subject matter of Example 164 or Example 165 may optionally include determining a first estimated subcutaneous temperature that is time-correlated with the first value and determining a second estimated subcutaneous temperature that is time-correlated with the second value, the second estimated subcutaneous temperature being determined at least in part based on a comparison of the second value to the first value.

In Example 167, the subject matter of Example 166 may optionally include determining a third estimated subcutaneous temperature time-correlated with the second value, determining whether a condition is met based on a comparison of the third estimated subcutaneous temperature to the second estimated subcutaneous temperature, and declaring an error or triggering a reset in response to the condition being met.

In Example 168, the subject matter of Example 167 may optionally include triggering a reset, where triggering the reset includes determining a subsequent estimated subcutaneous temperature based on the third estimated temperature and the second value, or based on the third conductance value and a fourth estimated subcutaneous temperature time-correlated with the third conductance value.

In Example 169, the subject matter of any one or any combination of Examples 164-168 may optionally include compensating for drift in the conductance value.

In Example 170, the subject matter of any one or any combination of Examples 164 to 169 may be optionally configured such that compensating for drift includes applying a filter.

One example (Example 171) of the subject matter (e.g., a method, system, or device) may include determining a first value indicative of a conductance of a sensor component at a first time, determining a second value indicative of a conductance of the sensor component at a later time, and determining an estimated subcutaneous temperature based at least in part on a comparison of the second value to the first value.

One example (Example 172) of the subject matter (e.g., a method, system, or device) may include accessing, by an analyte sensor system, first data from a system temperature sensor of the analyte sensor system, applying the first data to a trained temperature compensation model, the trained temperature compensation model to generate a compensated temperature value, and determining an estimated analyte value based at least in part on the compensated temperature value.

In example 173, the subject matter of example 172 may be configured such that the first data includes at least one of an uncompensated temperature value or raw temperature sensor data from a system temperature sensor.

In example 174, the subject matter of either example 172 or 173 may be configured such that the trained temperature compensation model may further include returning a first temperature sensor parameter in response to the first data and generating a compensated temperature value based at least in part on the first temperature sensor parameter.

In Example 175, the subject matter of any one or more of Examples 172-174 may be configured such that the trained temperature compensation model returns a system temperature sensor offset and a system temperature sensor slope, and further includes receiving raw sensor data from the system temperature sensor, and generating a compensated temperature value based at least in part on the raw sensor data, the system temperature sensor offset, and the system temperature sensor slope.

One example (Example 176) of the subject matter (e.g., a method, system, or device) may include an analyte sensor, a system temperature sensor, and a control circuit. The control circuit may be configured to perform operations including accessing first data from a system temperature sensor of an analyte sensor system, applying the first data to a trained temperature compensation model, the trained temperature compensation model to generate a compensated temperature value, and determining an estimated analyte value based at least in part on the compensated temperature value.

In example 177, the subject matter of example 176 may be configured such that the first data includes at least one of an uncompensated temperature value or raw temperature sensor data from a system temperature sensor.

In example 178, the subject matter of either example 176 or 177 may be configured such that the trained temperature compensation model may further include returning a first temperature sensor parameter in response to the first data and generating a compensated temperature value based at least in part on the first temperature sensor parameter.

In Example 179, the subject matter of any one or more of Examples 176-178 may be configured such that the trained temperature compensation model returns a system temperature sensor offset and a system temperature sensor slope, and may further include receiving raw sensor data from the system temperature sensor, and generating a compensated temperature value based at least in part on the raw sensor data, the system temperature sensor offset, and the system temperature sensor slope.

In example 180, any one or more of the subject matter of examples 176-179 may further include an application specific integrated circuit (ASIC) having a system temperature sensor.

One example (Example 181) of the subject matter (e.g., a method, system, or device) may include determining a temperature-compensated glucose concentration level. The determining may include receiving a glucose sensor signal, receiving a temperature parameter signal, detecting an exercise condition or state based at least in part on the glucose sensor signal or the temperature parameter signal, and modifying a temperature compensation applied to the glucose sensor signal.

In example 182, the subject matter of example 181 may include determining that the noise floor of the glucose sensor signal is greater than a first threshold.

In example 183, the subject matter of either example 181 or 182 may include determining that the noise floor of the temperature parameter signal is greater than a second threshold.

In Example 184, the subject matter of any one or more of Examples 181-183 may include determining that a noise floor of the glucose sensor signal is greater than a first threshold and determining that a noise floor of the temperature parameter signal is greater than a second threshold.

In example 185, the subject matter of any one or more of examples 181-184 may be configured such that modifying the temperature compensation includes applying a kinetic model to the temperature parameter signal to generate an estimated temperature parameter signal, and generating a temperature-compensated glucose concentration value using the estimated temperature parameter.

In example 186, the subject matter of any one or more of examples 181-185 may be configured such that detecting the exercise condition or state includes determining that a distribution of the rate of change of the temperature parameter signal satisfies a classifier.

In example 187, the subject matter of any one or more of examples 181-186 may be configured such that detecting the motion condition or state includes determining that the distribution of the rate of change of the temperature parameter signal is less than a threshold value.

One example (Example 188) of the subject matter (e.g., a method, system, or device) may include a temperature compensated glucose sensor system including a glucose sensor configured to generate a first signal representative of a glucose concentration in a host, a temperature sensor configured to generate a second signal representative of a temperature, and a processor. The processor may be programmed to perform operations including detecting an exercise condition or state based at least in part on the first signal or the second signal, and modifying a temperature compensation applied to the first signal.

In example 189, the subject matter of example 188 may be configured such that the operations further include determining that the noise floor of the first signal is greater than the first threshold.

In example 190, the subject matter of either example 188 or 189 may be configured such that the operation further includes determining that the noise floor of the second signal is greater than the second threshold.

In Example 191, the subject matter of any one or more of Examples 188-190 may be configured such that the operations further include determining that the noise floor of the first signal is greater than a first threshold and determining that the noise floor of the second signal is greater than a second threshold.

In Example 192, the subject matter of any one or more of Examples 188-191 may be configured such that modifying the temperature compensation includes applying a kinetic model to the second signal to generate an evaluated second signal, and generating a temperature-compensated glucose concentration value using the evaluated second signal.

In example 193, the subject matter of any one or more of examples 188-192 may be configured such that detecting the motion condition or state includes determining that the distribution of the rate of change of the second signal satisfies a classifier.

In Example 194, the subject matter of any one or more of Examples 188-193 may be configured such that detecting the motion condition or state includes determining that the distribution of the rate of change of the second signal is less than a threshold value.

One example ("Example 195") of the subject matter (e.g., methods, systems, or devices) may include a processor-implemented method of measuring temperature in an analyte sensor system. The method may include accessing a record of cyclical temperatures stored in the analyte sensor system during a first sensor session, determining a peak temperature from the record of cyclical temperatures, and performing a response action based on the peak temperature.

In example 196, the subject matter of example 195 may include determining that the peak temperature exceeds a peak temperature threshold, and the response action includes aborting the first sensor session.

In example 197, the subject matter of either example 195 or 196 may include determining initial sensor session parameters based at least in part on the peak temperature, receiving raw sensor data from an analyte sensor of the analyte sensor system, and generating an estimated analyte value using the initial session parameters and the raw sensor data.

In example 198, the subject matter of any one or more of examples 195-197 may be configured such that the initial sensor session parameters include sensitivity or baseline.

In example 199, the subject matter of any one or more of examples 195-198 may include, prior to a first sensor session, measuring a first temperature at the analyte sensor system, writing the first temperature to a periodic temperature log, waiting for a period, and measuring a second temperature at the analyte sensor system.

One example of the subject matter ("Example 200") may include a temperature compensated analyte sensor system. The temperature compensated analyte sensor system may include an analyte sensor configured to generate a first signal representative of an estimated analyte value in a host, a temperature sensor configured to generate a second signal representative of a temperature, and a processor. The processor may be programmed to perform operations including accessing a cyclical temperature record stored in the analyte sensor system during a first sensor session, determining a peak temperature from the cyclical temperature record, and performing a response action based on the peak temperature.

In example 201, the subject matter of example 200 may be configured such that the operation further includes determining that the peak temperature exceeds a peak temperature threshold and the response action includes aborting the first sensor session.

In example 202, the subject matter of either example 200 or 201 may be configured such that the operations further include determining initial sensor session parameters based at least in part on the peak temperature, receiving raw sensor data from an analyte sensor of the analyte sensor system, and generating an estimated analyte value using the initial session parameters and the raw sensor data.

In example 203, the subject matter of any one or more of examples 200-202 may be configured such that the initial sensor session parameters include sensitivity or baseline.

In example 204, the subject matter of any one or more of examples 200-203 may be configured such that the operations further include, prior to the first sensor session, measuring a first temperature at the analyte sensor system, writing the first temperature to the periodic temperature log, waiting for a period, and measuring a second temperature at the analyte sensor system.

One example ("Example 205") of the subject matter (e.g., a method, system, or device) may include a temperature sensitive analyte sensor system. The temperature sensitive analyte sensor system may include a diode, an electronic circuit, a sample and hold circuit, and a dual slope integrating analog-to-digital converter (ADC). The electronic circuit may be configured to perform operations including applying a first current to the diode for a first period, where a voltage drop across the diode has a first voltage value when the first current is provided to the diode, and applying a second current, different from the first current, to the diode for a second period after the first period, where a voltage drop across the diode has a second voltage value when the second current is provided to the diode. The sample and hold circuit may be configured to receive the first voltage value when the first voltage is applied to the diode and to generate an output indicative of the first voltage. A dual-slope integrating analog-to-digital converter (ADC) may include a first input coupled to receive a first voltage value from an output of the sample-and-hold circuit and a second input coupled to receive a voltage drop across a diode. The time for the output of the dual-slope integrating ADC to decay from the first voltage value to the second voltage value may be proportional to the temperature at the diode.

In example 206, the subject matter of example 205 may further include a comparator coupled to compare the output of the sample and hold circuit with the output of the dual slope integrating analog-to-digital circuit.

In example 207, the subject matter of either example 205 or 206 may further include a digital counter. The operations may further include initiating the digital counter at a peak of the output of the dual slope integrating ADC and determining a value of the digital counter in response to a change in the output of the comparator.

In example 208, the subject matter of any one or more of examples 205-207 may be configured such that a value of a digital counter indicating the time it takes for the output of the dual slope integrating ADC to decay from a first voltage value to a second voltage value is proportional to the temperature at the diode.

In example 209, the subject matter of any one or more of examples 205-208 may further include an AND circuit configured to generate a logical product between the output of the comparator and a clock signal, the clock signal being low when the first current is applied to the diode.

In example 210, the subject matter of any one or more of examples 205-209 may be configured such that the diode includes a diode-connected transistor.

In example 211, the subject matter of any one or more of examples 205-210 may be configured such that the analyte sensor of the analyte sensor is inserted into the skin of the host and the diode is positioned proximate to the skin of the host.

In Example 212, the subject matter of any one or more of Examples 205-211 may further include a first constant current source providing a first current, and a second pulsed current source, the second current including the sum of the first current and a current provided by the second pulsed current source when the second pulsed current source is on.

One example ("Example 213") of the subject matter (e.g., a method, system, or device) may include: applying a first current to a diode for a first period, where a voltage drop across the diode has a first voltage value when the first current is provided to the diode; applying a second current, different from the first current, to the diode after the first period, where a voltage drop across the diode has a second voltage value when the second current is provided to the diode; and providing the first and second voltage values to a dual-slope integrating analog-to-digital converter (ADC), where a time for an output of the dual-slope integrating ADC to decay from the first voltage value to the second voltage value is proportional to a temperature at the diode.

In example 214, the subject matter of example 213 may include comparing the output of the sample and hold circuit to the output of the dual slope integrating analog-to-digital circuit to generate a comparator output.

In example 215, the subject matter of either example 213 or 214 may include triggering a digital counter at a peak in the output of the dual slope integrating ADC and determining a value of the digital counter in response to a change in the comparator output.

In example 216, the subject matter of any one or more of examples 213-215 may be configured such that a value of a digital counter indicating the time it takes for the output of the dual slope integrating ADC to decay from a first voltage value to a second voltage value is proportional to the temperature at the diode.

In Example 217, the subject matter of any one or more of Examples 213-216 may include an AND circuit configured to generate a logical product between the output of the comparator and a clock signal. The clock signal may be low when the first current is applied to the diode.

In Example 218, the subject matter of any one or more of Examples 213-217 may be configured such that the diode includes a diode-connected transistor.

In example 219, the subject matter of any one or more of examples 213-218 can be configured such that the analyte sensor of the analyte sensor is inserted into the skin of the host and the diode is positioned proximate to the skin of the host.

One example ("Example 220") of the subject matter (e.g., methods, systems, or devices) may include a method of determining a glucose concentration level. The method may include receiving a temperature sensor signal, receiving a glucose sensor signal from a glucose sensor inserted at an insertion site in the host, and applying the temperature sensor signal and the glucose sensor signal to a model describing a difference between the glucose concentration at the insertion site and the blood glucose concentration in the host to generate a compensated blood glucose concentration for the host.

In example 221, the subject matter of example 220 may include determining a model time parameter based at least in part on the temperature sensor signal, and determining a compensated blood glucose concentration based at least in part on the model time parameter.

In example 222, the subject matter of either example 220 or 221 can be configured such that the model time parameters are applied to the glucose concentration at the insertion site and to the blood glucose concentration.

In example 223, the subject matter of any one or more of examples 220-222 can further include determining glucose consumption describing the host. The compensated blood glucose concentration can be based at least in part on the glucose consumption.

In Example 224, the subject matter of any one or more of Examples 220-223 may further include determining glucose consumption using a constant cell layer glucose concentration.

In Example 225, the subject matter of any one or more of Examples 220-224 may further include determining glucose consumption using the variable cell layer glucose concentration.

In Example 226, the subject matter of any one or more of Examples 220-225 may include determining glucose consumption using a linearly varying cell layer glucose concentration.

One example ("Example 227") of the subject matter (e.g., a method, system, or device) may include a temperature compensated glucose sensor system. The temperature compensated glucose sensor system may include a glucose sensor and sensor electronics. The sensor electronics may be configured to perform operations including receiving a temperature sensor signal, receiving a glucose sensor signal from a glucose sensor inserted at an insertion site in a host, and applying the temperature sensor signal and the glucose sensor signal to a model describing the difference between the glucose concentration at the insertion site and the blood glucose concentration in the host to generate a compensated blood glucose concentration for the host.

In example 228, the subject matter of example 227 is configured such that the operations further include determining a model time parameter based at least in part on the temperature sensor signal, and determining a compensated blood glucose concentration based at least in part on the model time parameter.

In Example 229, the subject matter of either Example 227 or 228 can be configured such that the model time parameters are applied to the glucose concentration at the insertion site and to the blood glucose concentration.

In example 230, the subject matter of any one or more of examples 227-229 may be configured such that the operations further include determining glucose consumption describing the host, and the compensated blood glucose concentration is based at least in part on the glucose consumption.

In Example 231, the subject matter of any one or more of Examples 227-230 may be configured such that the operations further include determining glucose consumption using a constant cell layer glucose concentration.

In Example 232, the subject matter of any one or more of Examples 227-230 may be configured such that the operations further include determining glucose consumption using the variable cell layer glucose concentration.

In Example 233, the subject matter of any one or more of Examples 227-231 may be configured such that the operations further include determining glucose consumption using the linearly varying cell layer glucose concentration.

Example 234 is an analyte sensor system for generating an estimated analyte value, comprising an in vitro temperature sensor, an in vivo analyte sensor, and at least one processor, the analyte sensor system being programmed to perform operations including accessing a first sensor signal from the in vivo analyte sensor, accessing a first temperature signal from the in vitro temperature sensor, generating a first analyte sensor temperature based at least in part on the first temperature signal, determining that the first analyte sensor temperature satisfies a temperature condition, generating a first temperature compensated sensitivity based at least in part on the first analyte sensor temperature, and generating a first estimated analyte value based at least in part on the first sensor signal and the first temperature compensated sensitivity.

In example 235, the subject matter of example 234 optionally includes determining that the first analyte sensor temperature satisfies the temperature condition includes determining that the first analyte sensor temperature is within a first temperature range.

In Example 236, the subject matter of either Example 234 or 235 optionally includes determining that the first analyte sensor temperature satisfies the temperature condition includes generating a temperature rate of change using the first analyte sensor temperature and at least one previous analyte sensor temperature indicated by the previous temperature signal, and determining that the temperature rate of change satisfies the rate of change condition.

In example 237, the subject matter of any one or more of examples 234-236 includes an operation that optionally further includes generating a first in vitro temperature using the first temperature signal and generating a first analyte sensor temperature using the first in vitro temperature.

In Example 238, the subject matter of Example 237 optionally includes determining that the first in vitro temperature indicated by the first temperature signal satisfies the temperature condition includes determining that a difference between the first in vitro temperature and the first analyte sensor temperature is less than a threshold value.

In example 239, the subject matter of any one or more of examples 234-238 includes an operation that optionally further includes determining that the second analyte sensor temperature indicated by the second temperature signal from the in vitro temperature sensor does not meet the temperature condition, and performing a response action in response to determining that the second analyte sensor temperature indicated by the second temperature signal does not meet the temperature condition.

In example 240, the subject matter of example 239 optionally includes responsive actions including generating a second temperature compensated sensitivity based at least in part on the default analyte sensor temperature and generating a second estimated analyte value based at least in part on the second analyte sensor signal and the second temperature compensated sensitivity.

In example 241, the subject matter of either example 239 or 240 optionally includes a responsive action that includes pausing display of the analyte value on a display associated with the analyte sensor.

In example 242, the subject matter of any one or more of examples 239-241 optionally includes a responsive action that includes generating a second estimated analyte value based at least in part on the second analyte sensor signal and the temperature uncompensated sensitivity.

In Example 243, the subject matter of any one or more of Examples 234-242 optionally includes operations further including receiving a second sensor signal from the in vivo analyte sensor, generating a second temperature compensated sensitivity based at least in part on the first analyte sensor temperature, and generating a second estimated analyte value based at least in part on the second sensor signal and the second temperature compensated sensitivity.

In Example 244, the subject matter of any one or more of Examples 234-243 optionally includes an operation further including determining that the first temperature signal satisfies a temperature signal condition before generating the first temperature compensated sensitivity.

In example 245, the subject matter of example 244 optionally includes determining that the first temperature signal satisfies the temperature signal condition being performed by an in vitro temperature sensor.

Example 246 is a method of generating an estimated analyte value, the method including: accessing a first sensor signal from an in vivo analyte sensor; accessing a first temperature signal from an in vitro temperature sensor; generating a first analyte sensor temperature based at least in part on the first temperature signal; determining that the first analyte sensor temperature satisfies a temperature condition; generating a first temperature compensated sensitivity based at least in part on the first analyte sensor temperature; and generating a first estimated analyte value based at least in part on the first sensor signal and the first temperature compensated sensitivity.

In Example 247, the subject matter of Example 246 optionally includes determining that the first analyte sensor temperature satisfies the temperature condition includes determining that the first analyte sensor temperature is within a first temperature range.

In Example 248, the subject matter of either Example 246 or 247 optionally includes determining that the first analyte sensor temperature satisfies the temperature condition includes generating a temperature rate of change using the first analyte sensor temperature and at least one previous analyte sensor temperature indicated by the previous temperature signal, and determining that the temperature rate of change satisfies the rate of change condition.

In Example 249, the subject matter of any one or more of Examples 246-248 optionally includes generating a first in vitro temperature using the first temperature signal and generating a first analyte sensor temperature using the first in vitro temperature.

In Example 250, the subject matter of Example 249 optionally includes determining that the first in vitro temperature indicated by the first temperature signal satisfies the temperature condition includes determining that a difference between the first in vitro temperature and the first analyte sensor temperature is less than a threshold value.

In example 251, the subject matter of any one or more of examples 246-250 optionally includes determining that the second analyte sensor temperature indicated by the second temperature signal does not meet the temperature condition, and performing a response action in response to determining that the second analyte sensor temperature indicated by the second temperature signal does not meet the temperature condition.

In Example 252, the subject matter of Example 251 optionally includes responsive actions including generating a second temperature compensated sensitivity based at least in part on the default analyte sensor temperature and generating a second estimated analyte value based at least in part on the second analyte sensor signal and the second temperature compensated sensitivity.

In example 253, the subject matter of either example 251 or 252 optionally includes a responsive action that includes pausing display of the analyte value on a display associated with the analyte sensor.

In example 254, the subject matter of any one or more of examples 251-253 optionally includes a responsive action that includes generating a second estimated analyte value based at least in part on the second analyte sensor signal and the temperature uncompensated sensitivity.

In example 255, the subject matter of any one or more of examples 246-254 optionally includes receiving a second sensor signal from the in vivo analyte sensor, generating a second temperature compensated sensitivity based at least in part on the first analyte sensor temperature, and generating a second estimated analyte value based at least in part on the second sensor signal and the second temperature compensated sensitivity.

In Example 256, the subject matter of any one or more of Examples 246-255 optionally includes determining that the first temperature signal satisfies a temperature signal condition prior to generating the first temperature compensated sensitivity.

In example 257, the subject matter of example 256 optionally includes determining that the first temperature signal satisfies the temperature signal condition being performed by an in vitro temperature sensor.

Example 258 is an analyte sensor system for generating an estimated analyte value, comprising an in vitro temperature sensor, an in vivo analyte sensor, and at least one processor, the method including operations of accessing a first sensor signal from the in vivo analyte sensor, accessing a first temperature signal from the in vitro temperature sensor, selecting a first temperature compensation model from a set of temperature compensation models, the selecting being based at least in part on the first temperature signal, generating a first analyte sensor temperature based at least in part on the first temperature signal and the first temperature compensation model, generating a first temperature compensated sensitivity based at least in part on the first analyte sensor temperature, and generating a first estimated analyte value based at least in part on the first sensor signal and the first temperature compensated sensitivity.

In Example 259, the subject matter of Example 258 optionally includes a set of temperature compensation models including piecewise linear models, where a first temperature compensation model corresponds to a first segment of the piecewise linear model over a first temperature range and a second temperature compensation model corresponds to a second segment of the piecewise linear model over a second temperature range different from the first temperature range.

In example 260, the subject matter of either example 258 or 259 optionally includes an operation further including determining a rate of change of temperature of the in vivo analyte sensor using the first temperature signal, the first temperature compensation model corresponding to the rate of change of temperature.

In example 261, the subject matter of any one or more of examples 258-260 optionally includes an operation further including generating a difference between the in vitro temperature indicated by the in vitro temperature sensor and the first analyte sensor temperature, the first temperature compensation model corresponding to the difference.

In Example 262, the subject matter of any one or more of Examples 258-261 optionally includes the first temperature compensation model having a first linear gain, and the set of temperature compensation models also including a second temperature compensation model having a second linear gain different from the first linear gain.

In Example 263, the subject matter of any one or more of Examples 258-262 optionally includes the set of temperature compensation models including a first temperature compensation model and a second temperature compensation model, and at least one of the first temperature compensation model or the second temperature compensation model is a constant.

In example 264, the subject matter of any one or more of examples 258-263 includes operations further including accessing a second sensor signal from the in vivo analyte sensor, accessing a second temperature signal from the in vitro temperature sensor, generating a second analyte sensor temperature based at least in part on the second temperature signal, determining that the second analyte sensor temperature does not meet the temperature condition, and performing a response action in response to determining that the second analyte sensor temperature does not meet the temperature condition.

In Example 265, the subject matter of Example 264 optionally includes responsive actions including generating a second temperature compensated sensitivity based at least in part on the default analyte sensor temperature and generating a second estimated analyte value based at least in part on the second sensor signal and the second temperature compensated sensitivity.

In example 266, the subject matter of either example 264 or 265 optionally includes a responsive action that includes pausing display of the analyte value on a display associated with the analyte sensor.

In example 267, the subject matter of any one or more of examples 264-266 optionally includes a responsive action that includes generating a second estimated analyte value based at least in part on the second sensor signal and the temperature uncompensated sensitivity.

In Example 268, the subject matter of any one or more of Examples 258-267 optionally includes operations further including receiving a second sensor signal from the in vivo analyte sensor, generating a second temperature compensated sensitivity based at least in part on the first analyte sensor temperature, and generating a second estimated analyte value based at least in part on the second sensor signal and the second temperature compensated sensitivity.

In example 269, the subject matter of any one or more of examples 258-268 optionally includes an operation further including determining that the first temperature signal satisfies a temperature signal condition prior to generating the first temperature compensated sensitivity.

In example 270, the subject matter of example 269 optionally includes determining that the first temperature signal satisfies the temperature signal condition being performed by an in vitro temperature sensor.

Example 271 is a method of generating an estimated analyte value, the method including: accessing a first sensor signal from an in vivo analyte sensor; accessing a first temperature signal from an in vitro temperature sensor; selecting a first temperature compensation model from a set of temperature compensation models, where selecting is based at least in part on the first temperature signal; generating a first analyte sensor temperature based at least in part on the first temperature signal and the first temperature compensation model; generating a first temperature compensated sensitivity based at least in part on the first analyte sensor temperature; and generating a first estimated analyte value based at least in part on the first sensor signal and the first temperature compensated sensitivity.

In Example 272, the subject matter of Example 271 optionally includes a set of temperature compensation models including piecewise linear models, where a first temperature compensation model corresponds to a first segment of the piecewise linear model over a first temperature range and a second temperature compensation model corresponds to a second segment of the piecewise linear model over a second temperature range different from the first temperature range.

In Example 273, the subject matter of either Example 271 or 272 optionally includes determining a rate of change of temperature of the in vivo analyte sensor using the first temperature signal, and the first temperature compensation model corresponds to the rate of temperature change.

In example 274, the subject matter of any one or more of examples 271-273 optionally includes generating a difference between the in vitro temperature indicated by the in vitro temperature sensor and the first analyte sensor temperature, and the first temperature compensation model corresponds to the difference.

In Example 275, the subject matter of any one or more of Examples 271-274 optionally includes the first temperature compensation model having a first linear gain, and the set of temperature compensation models also includes a second temperature compensation model having a second linear gain different from the first linear gain.

In Example 276, the subject matter of any one or more of Examples 271-275 optionally includes the set of temperature compensation models comprising a first temperature compensation model and a second temperature compensation model, and at least one of the first temperature compensation model or the second temperature compensation model is a constant.

In Example 277, the subject matter of any one or more of Examples 271-276 optionally includes accessing a second sensor signal from the in vivo analyte sensor, accessing a second temperature signal from the in vitro temperature sensor, generating a second analyte sensor temperature based at least in part on the second temperature signal, determining that the second analyte sensor temperature does not meet the temperature condition, and performing a response action in response to determining that the second analyte sensor temperature does not meet the temperature condition.

In Example 278, the subject matter of Example 277 optionally includes responsive actions including generating a second temperature compensated sensitivity based at least in part on the default analyte sensor temperature and generating a second estimated analyte value based at least in part on the second sensor signal and the second temperature compensated sensitivity.

In Example 279, the subject matter of either Example 277 or 278 optionally includes a responsive action that includes pausing display of the analyte value on a display associated with the analyte sensor.

In example 280, the subject matter of any one or more of examples 277-279 optionally includes a responsive action that includes generating a second estimated analyte value based at least in part on the second sensor signal and the temperature uncompensated sensitivity.

In Example 281, the subject matter of any one or more of Examples 271-280 optionally includes receiving a second sensor signal from the in vivo analyte sensor, generating a second temperature compensated sensitivity based at least in part on the first analyte sensor temperature, and generating a second estimated analyte value based at least in part on the second sensor signal and the second temperature compensated sensitivity.

In Example 282, the subject matter of any one or more of Examples 271-281 optionally includes determining that the first temperature signal satisfies a temperature signal range before generating the first temperature compensated sensitivity.

In Example 283, the subject matter of Example 282 optionally includes determining that the first temperature signal satisfies a temperature signal range, performed by an in vitro temperature sensor.

Example 284 is a glucose sensor system for generating an estimated glucose value, comprising an in vitro temperature sensor, an in vivo analyte sensor, and at least one processor, the glucose sensor system being programmed to perform operations including accessing a first sensor signal from the in vivo glucose sensor, accessing a first temperature signal from the in vitro temperature sensor, generating a first glucose sensor temperature based at least in part on the first temperature signal, generating a first temperature-compensated sensitivity based at least in part on the first glucose sensor temperature, generating a first temperature-compensated non-glucose signal based at least in part on the first glucose sensor temperature, and generating a first estimated glucose value based at least in part on the first sensor signal, the first temperature-compensated sensitivity, and the first temperature-compensated non-glucose signal.

In Example 285, the subject matter of Example 284 includes operations that optionally further include accessing a second sensor signal from the in vivo glucose sensor, accessing a second temperature signal from the in vitro temperature sensor, generating a second glucose sensor temperature based at least in part on the second temperature signal, determining that the second glucose sensor temperature does not satisfy the temperature condition, generating a second temperature-compensated non-glucose signal based at least in part on a default temperature, and generating a second estimated glucose value based at least in part on the second sensor signal and the second temperature-compensated non-glucose signal.

In Example 286, the subject matter of either Example 284 or 285 optionally includes where generating the first temperature compensated non-glucose signal includes determining a difference between the first glucose sensor temperature and a reference glucose sensor temperature and applying a non-glucose compensation factor to the difference.

In example 287, the subject matter of any one or more of examples 284-286 optionally includes an operation further including determining that the first glucose sensor temperature satisfies a temperature condition before generating the first temperature compensated sensitivity.

In Example 288, the subject matter of Example 287 optionally includes determining that the first glucose sensor temperature satisfies the temperature condition includes determining that the first glucose sensor temperature is within a first temperature range.

In Example 289, the subject matter of either Example 287 or 288 optionally includes determining that the first glucose sensor temperature satisfies the temperature condition includes generating a temperature rate of change using the first glucose sensor temperature and at least one previous glucose sensor temperature indicated by the previous temperature signal, and determining that the temperature rate of change does not exceed the rate of change condition.

In example 290, the subject matter of any one or more of examples 287-289 includes an operation that optionally further includes generating a first in vitro temperature using the first temperature signal and generating a first glucose sensor temperature using the first in vitro temperature.

In Example 291, the subject matter of Example 290 optionally includes determining that the first in vitro temperature indicated by the first temperature signal satisfies the temperature condition includes determining that a difference between the first in vitro temperature and the first glucose sensor temperature is less than a threshold value.

In Example 292, the subject matter of any one or more of Examples 284-291 optionally includes operations further including: accessing a second sensor signal from the in vivo glucose sensor; accessing a second temperature signal from the in vitro temperature sensor; generating a second glucose sensor temperature based at least in part on the second temperature signal; determining that the second glucose sensor temperature does not satisfy the temperature condition; and performing a response action in response to determining that the second glucose sensor temperature exceeds the temperature condition.

In Example 293, the subject matter of Example 292 optionally includes responsive actions including generating a second temperature compensated sensitivity based at least in part on the default glucose sensor temperature, and generating a second estimated glucose value based at least in part on the second sensor signal and the second temperature compensated sensitivity.

In Example 294, the subject matter of either Example 292 or 293 optionally includes a responsive action that includes pausing the display of the glucose concentration on a display associated with the glucose sensor.

In example 295, the subject matter of any one or more of examples 292-294 optionally includes a responsive action that includes generating a second estimated glucose value based at least in part on the second sensor signal and the uncompensated sensitivity.

In Example 296, the subject matter of any one or more of Examples 284-295 optionally includes operations further including receiving a second sensor signal from the in vivo glucose sensor, generating a second temperature compensated sensitivity based at least in part on the first glucose sensor temperature, and generating a second estimated glucose value based at least in part on the second sensor signal and the second temperature compensated sensitivity.

In example 297, the subject matter of any one or more of examples 284-296 optionally includes an operation further including determining that the first temperature signal satisfies a temperature signal condition prior to generating the first temperature compensated sensitivity.

In example 298, the subject matter of example 297 optionally includes that determining that the first temperature signal satisfies the temperature signal condition is performed by an in vitro temperature sensor.

Example 299 is a method for generating an estimated glucose value, the method including: accessing a first sensor signal from an in vivo glucose sensor; accessing a first temperature signal from an in vitro temperature sensor; generating a first glucose sensor temperature based at least in part on the first temperature signal; generating a first temperature compensated sensitivity based at least in part on the first glucose sensor temperature; generating a first temperature compensated non-glucose signal based at least in part on the first glucose sensor temperature; and generating a first estimated glucose value based at least in part on the first sensor signal, the first temperature compensated sensitivity, and the first temperature compensated non-glucose signal.

In example 300, the subject matter of example 299 optionally includes accessing a second sensor signal from the in vivo glucose sensor, accessing a second temperature signal from the in vitro temperature sensor, generating a second glucose sensor temperature based at least in part on the second temperature signal, determining that the second glucose sensor temperature does not satisfy a temperature condition, generating a second temperature-compensated non-glucose signal based at least in part on a default temperature, and generating a second estimated glucose value based at least in part on the second sensor signal and the second temperature-compensated non-glucose signal.

In Example 301, the subject matter of either Example 299 or 300 optionally includes where generating the first temperature compensated non-glucose signal includes determining a difference between the first glucose sensor temperature and a reference glucose sensor temperature and applying a non-glucose compensation factor to the difference.

In example 302, the subject matter of any one or more of examples 299-301 optionally includes determining that the first glucose sensor temperature satisfies a temperature condition before generating the first temperature compensated sensitivity.

In example 303, the subject matter of example 302 optionally includes determining that the first glucose sensor temperature satisfies the temperature condition includes determining that the first glucose sensor temperature is within a first temperature range.

In Example 304, the subject matter of either Example 302 or 303 optionally includes determining that the first glucose sensor temperature satisfies the temperature condition includes generating a temperature change rate using the first glucose sensor temperature and at least one previous glucose sensor temperature indicated by the previous temperature signal, and determining that the temperature change rate does not exceed the change rate condition.

In example 305, the subject matter of any one or more of examples 302-304 optionally includes generating a first in vitro temperature using the first temperature signal and generating a first glucose sensor temperature using the first in vitro temperature.

In Example 306, the subject matter of Example 305 optionally includes determining that the temperature indicated by the first temperature signal satisfies the temperature condition includes determining that a difference between the first ex vivo temperature and the first glucose sensor temperature is less than a threshold value.

In example 307, the subject matter of any one or more of examples 299-306 optionally includes accessing a second sensor signal from the in vivo glucose sensor, accessing a second temperature signal from the in vitro temperature sensor, generating a second glucose sensor temperature based at least in part on the second temperature signal, determining that the second glucose sensor temperature does not satisfy the temperature condition, and performing a response action in response to determining that the second glucose sensor temperature exceeds the temperature condition.

In example 308, the subject matter of example 307 optionally includes responsive actions including generating a second temperature compensated sensitivity based at least in part on the default glucose sensor temperature, and generating a second estimated glucose value based at least in part on the second sensor signal and the second temperature compensated sensitivity.

In example 309, the subject matter of either example 307 or 308 optionally includes a responsive action that includes pausing the display of the glucose concentration on a display associated with the glucose sensor.

In example 310, the subject matter of any one or more of examples 307-309 optionally includes a responsive action that includes generating a second estimated glucose value based at least in part on the second sensor signal and the temperature uncompensated sensitivity.

In Example 311, the subject matter of any one or more of Examples 299-310 optionally includes receiving a second sensor signal from the in vivo glucose sensor, generating a second temperature compensated sensitivity based at least in part on the first glucose sensor temperature, and generating a second estimated glucose value based at least in part on the second sensor signal and the second temperature compensated sensitivity.

In Example 312, the subject matter of any one or more of Examples 299-311 optionally includes determining that the first temperature signal satisfies a temperature signal condition prior to generating the first temperature compensated sensitivity.

In example 313, the subject matter of example 312 optionally includes determining that the first temperature signal satisfies the temperature signal condition being performed by an in vitro temperature sensor.

An example (e.g., "Example 314") of the subject matter (e.g., a system or apparatus) may optionally combine any part of any one or more of Examples 1-313 or any combination of any part of any one or more of Examples 1-313 to include a "means for" performing any part of any one or more of the functions or methods of Examples 1-313, or a "machine-readable medium" (e.g., a collective machine-readable medium, a non-transitory machine-readable medium, etc.) that includes instructions that, when executed by a machine, cause a machine to perform any part of any one or more of the functions or methods of Examples 1-313.

This Summary is intended to provide an overview of the subject matter of the present patent application. It is not intended to provide an exclusive or exhaustive description of the present disclosure. The Detailed Description is included to provide further information about the present patent application. Other aspects of the present disclosure will be apparent to those of ordinary skill in the art upon reading and understanding the following Detailed Description and viewing the drawings that form a part hereof, each of which is not to be construed in a limiting sense.

In the drawings, which are not necessarily drawn to scale, like numerals may describe like components in different figures. Like numerals with different letter suffixes may represent different instances of the like components. The drawings illustrate generally, by way of example, and not by way of limitation, various embodiments discussed in the present document.
FIG. 1 is a diagram of an exemplary analyte sensor system that may include a temperature sensor and in which a temperature compensation method may be implemented. 1 is a schematic diagram of an exemplary analyte sensor system. FIG. 2 is a schematic diagram of an exemplary sensor electronics portion of the analyte sensor system. FIG. 1 is a schematic diagram of an exemplary analyte sensor system engaged with tissue of a host. FIG. 1 is a schematic diagram of an exemplary analyte sensor system engaged with tissue of a host. FIG. 13 is a schematic diagram of a temperature sensor on a distal portion of the analyte sensor. FIG. 2 is a schematic diagram of an exemplary temperature sensor on a proximal portion of the analyte sensor. FIG. 13 is a schematic diagram of another exemplary temperature sensor on a proximal portion of the analyte sensor. FIG. 5B is an enlarged view of a portion of the temperature sensor shown in FIG. 5A. FIG. 11 is a flow chart diagram of an exemplary method for determining a temperature compensated glucose concentration level using a delay parameter. FIG. 1 is a flow chart diagram of an exemplary method for determining a temperature-compensated glucose concentration level based on an evaluated (e.g., confirmed) temperature value. 1 is a schematic diagram of an exemplary method for temperature compensating a continuous glucose sensor that includes determining a pattern from temperature information. FIG. 11 is a flow chart diagram of an exemplary method for temperature compensating a continuous glucose monitoring system based at least in part on a detected condition or state. 1 is a schematic diagram of a method for temperature compensating a continuous glucose sensor system using a reference temperature value. FIG. 1 is a flow chart diagram of an exemplary continuous glucose sensor temperature compensation method. FIG. 4 is a flow chart diagram of an exemplary method of temperature compensation using two temperature sensors. FIG. 1 is a flow chart diagram of an exemplary method for determining when continuous glucose (or other analyte) monitoring has been restarted. FIG. 1 is a flow chart diagram of an exemplary method for determining an anatomical location of a sensor. 1 shows the output of a glucose sensor plotted against time. 1 shows the output of a temperature sensor plotted against time. 1 shows the overlay of temperature versus glucose sensor output, where the correlation is apparent. Shown is a superposition of temperature versus glucose sensor output, where no correlation is apparent. FIG. 13 is a graph showing plots of temperature versus time for a sensor on the host's abdomen and a sensor on the host's arm. 1 is a plot of standard deviation versus mean temperature over the first 24 hours for several sensor devices. 1 is a plot of temperature versus time where the sensor electronics package was removed from the sensor for 1 minute. 1 is a plot of temperature versus time where the sensor electronics package was removed from the sensor for 5 minutes. FIG. 1 is a schematic diagram of an example model that may be used to determine an output from two or more inputs that may be received at different times. FIG. 4 is a flow chart diagram of an exemplary method for determining a compensated glucose concentration value using a model. FIG. 11 is a flow chart diagram of another exemplary method for determining a compensated glucose concentration value using a model. 1 is a graph showing temperature and impedance plotted against time. FIG. 13 is a flow chart diagram of an exemplary method of temperature compensation using conductance or impedance. FIG. 1 is a flow chart diagram of an exemplary method for determining an estimated subcutaneous temperature using conductance or impedance. FIG. 1 is a flow chart diagram of an exemplary method for training a temperature compensation model. FIG. 13 is a flowchart diagram of an exemplary method for utilizing a trained temperature compensation model. FIG. 1 is a flow chart diagram of an exemplary method for detecting an exercise condition or state. 1 is a graph showing a first change distribution function indicative of a host in a resting (eg, non-exercising) condition or state, and a second change distribution function indicative of a host in an exercise condition or state. FIG. 1 is a flow chart diagram of an exemplary method for detecting a motion condition or state using a distribution of rates of change in temperature parameter signal samples. FIG. 13 is a flow chart diagram of an exemplary method for recording temperature in an analyte sensor system during shipment. FIG. 13 is a flow chart diagram of an example method for initiating an analyte sensor session, the analyte sensor session including recording periodic temperature measurements obtained from shipping and/or storage of the analyte sensor system. FIG. 1 is a diagram of an example circuit configuration that may be implemented in an analyte sensor system to measure temperature using a diode. FIG. 1 is a flow chart diagram of a method for measuring temperature in an analyte sensor system using a diode. 1 illustrates an exemplary sensor insertion site showing a layer of cells between the sensor insertion site and a host capillary site. FIG. 1 illustrates an example of a process flow that may be implemented in an analyte sensor system that uses temperature compensation to respond to exceptions. FIG. 2 is a block diagram illustrating a temperature model. 1 shows a chart illustrating an example relationship between in-vitro temperature T _Tx and in-vivo sensor temperature T _WE , showing the range of in-vivo sensor temperatures T _WE that will trigger an exception. FIG. 37 shows a version of the chart of FIG. 36 that includes regions corresponding to situations when the in vitro temperature T _Tx is above an upper temperature limit and/or below a lower temperature limit. 11 is a flowchart illustrating an example of a process flow that may be performed by an analyte sensor system to determine an in vivo sensor temperature from an in vitro temperature measurement, taking into account model parameters. 13 is a flow chart illustrating an example of detecting an exception based on a determined in-vivo sensor temperature _TWE . 1 shows a chart illustrating an example relationship between ex-vivo temperature T _Tx and in-vivo sensor temperature T _WE . 1 is a flow chart illustrating an example of a process flow that may be performed by an analyte sensor system to use multiple temperature models over different ranges of in-vivo sensor temperatures _TWE . 42 is a flow chart illustrating another example of a process flow 4200 that may be performed by an analyte sensor system to use multiple temperature models over different ranges of in-vivo sensor temperatures _TWE . 1 is a flow chart illustrating an example of a process flow that may be performed by an analyte sensor system, such as a glucose sensor system, to determine a temperature compensated estimated glucose value. 1 is a flow chart illustrating an example of a process flow that may be performed by an analyte sensor system to upsample a signal from an in vitro temperature sensor. FIG. 1 is a schematic diagram of another exemplary analyte sensor system engaged with tissue of a host. FIG. 46 is a cross-sectional view of the exemplary configuration of FIG. 45 along line AA.

The accuracy of a glucose sensor is important to patients, caregivers, and clinicians because the estimated glucose values obtained from the glucose sensor may be used to determine treatment or evaluate treatment effectiveness. Several factors may affect the accuracy of a glucose sensor. One factor is temperature. The inventors recognize that, among other things, steps may be taken to compensate for the effects of temperature on the glucose sensor, thereby improving the performance of the sensor system by improving the accuracy of the estimated glucose level, thereby reducing the mean absolute relative deviation (MARD) of the sensor system. The MARD value over the useful or indicated range of glucose levels is a common way to describe the precision and accuracy of glucose measurements by a glucose sensing system. The MARD is the result of a mathematical calculation that measures the average disparity between the estimated glucose value produced by the glucose sensor and a reference measurement. The lower the MARD, the more accurate the device is considered to be.

DEFINITIONS To facilitate understanding of the various examples, some additional terms are defined below.

As used herein, the term "about" is a broad term that is to be given its ordinary and customary meaning to those of ordinary skill in the art (and is not limited to any special or customized meaning), and when associated with any numerical value or range, refers to, but is not limited to, the understanding that the amount or condition that it modifies may vary somewhat beyond the stated amount so long as the functionality of the embodiment is achieved.

As used herein, the term "A/D converter" is a broad term given its ordinary and customary meaning to those skilled in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, hardware and/or software that converts an analog signal into a corresponding digital signal.

As used herein, the term "analyte" is a broad term and is given its ordinary and customary meaning to one of ordinary skill in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, a substance or chemical component in a biological fluid (e.g., blood, interstitial fluid, cerebrospinal fluid, lymphatic fluid, or urine) that may be analyzed. Analytes may include naturally occurring substances, man-made substances, metabolites, and/or reaction products. In some embodiments, the analyte for measurement by the sensor heads, devices, and methods disclosed herein is glucose. However, other analytes are contemplated as well, including lactate; bilirubin; ketones; carbon dioxide; sodium; potassium; acarboxyprothrombin; acylcarnitines; adenine phosphoribosyltransferase; adenosine deaminase; albumin; α-fetoprotein; amino acid profile (arginine (Krebs cycle), histidine/urocanic acid, homocysteine, phenylalanine/tyrosine, tryptophan); andrenostenedione; antipyrine; arabinitol enantiomers; arginase; benzoylecgonine (cocaine); biotinidase; biopterin; c-reactive protein; carnitine; carnosinase; CD4; ceruloplasmin; chenodeoxycholic acid; chloroquine; cholesterol; cholinesterase; conjugated 1-β hydroxy-cholic acid; cortisol; creatine kinase; creatine kinase; tin kinase MM isoenzyme; cyclosporine A; d-penicillamine; deethylchloroquine; dehydroepiandrosterone sulfate; DNA (acetylation polymorphism, alcohol dehydrogenase, α1-antitrypsin, cystic fibrosis, Duchenne/Becker muscular dystrophy, analyte-6-phosphate dehydrogenase, hemoglobinopathies, A,S,C,E,D-Punjab, β-thalassemia, hepatitis B virus, HCMV, HI V-1, HTLV-1, Leber's hereditary optic neuropathy, MCAD, RNA, PKU, Plasmodium vivax, sex differentiation, 21-deoxycortisol); desbutylhalofantrine; dihydropteridine reductase; diphtheria/tetanus antitoxin; erythrocyte arginase; erythrocyte protoporphyrin; esterase D; fatty acid/acylglycine; free β-human chorionic gonadotropin; free erythrocyte porphyrin; free thyroxine (FT4); free triiodothyronine (free tri-iodothyronine, FT3); fumarylacetoacetase; galactose/gal-1-phosphate; galactose-1-phosphate uridyltransferase; gentamicin; analyte-6-phosphate dehydrogenase; glutathione; glutathione peroxidase; glycocholic acid; glycosylated hemoglobin; halofantrine; hemoglobin variants; hexosaminidase A; human erythrocyte carbonic anhydrase I; 17α-hydroxyprogesterone; hypoxanthine phosphoribosyltransferase; immunoreactive trypsin; lead; lipoproteins ((a), B/A-1, β); lysozyme; mefloquine; netilmicin; phenobarbitone; phenytoin; phytanic acid/pristanic acid; progesterone; prolactin; prolidase; purine nucleoside phosphorylase; quinine; reverse triiodothyronine tri-iodothyronine, rT3); selenium; serum pancreatic lipase; sisomicin; somatomedin C; specific antibodies (adenovirus, antinuclear antibody, anti-zeta antibody, arbovirus, pseudorabies virus, dengue virus, dracunculiasis, tapeworm, granulosus echinococcus, ameba histolytica, enterovirus, giardiasis, Helicobacter pylori, hepatitis B virus, herpes virus, HIV-1, IgE (atopic disease), influenza virus, Leishmania donovani, Leptospirosis, measles/mumps/rubella, Mycobacterium leprae, Mycoplasma pneumoniae, myoglobin, Onchocerca volvulus, parainfluenza virus, malarial Examples of analytes include Plasmodium, poliovirus, Pseudomonas aeruginosa, respiratory syncytial virus, Rickettsia (tsutsugamushi disease), Schistosoma mansoni, Toxoplasma gondii, Treponema pallidum, Trypanosoma cruzi/rangeli, Vesicular stomatitis virus, Wuchereria bancrofti, Yellow fever virus); specific antigens (Hepatitis B virus, HIV-1); succinylacetone; sulfadoxine; theophylline; thyrotropin (TSH); thyroxine (T4); thyroxine-binding globulin; trace elements; transferrin; UDP-galactose-4-epimerase; urea; uroporphyrinogen I synthase; vitamin A; white blood cells; and zinc protoporphyrin. Salts, sugars, proteins, fats, vitamins, and hormones naturally present in blood or interstitial fluids may also constitute analytes in certain embodiments. Analytes may occur naturally in biological fluids, such as metabolic products, hormones, antigens, antibodies, and the like. Alternatively, the analyte may be introduced into the body, for example a contrast agent for imaging, a radioisotope, a chemical agent, a fluorocarbon-based synthetic blood, or a drug or pharmaceutical composition, including, but not limited to, insulin; ethanol; cannabis (marijuana, tetrahydrocannabinol, hashish); inhalants (nitrous oxide, amyl nitrite, butyl nitrite, chlorohydrocarbons, hydrocarbons); cocaine (crack cocaine); stimulants (amphetamine, methamphetamine, ritalin, silurt, preludine, didrex, prestate, borane, sandrex, pregin); antidepressants (barbiturates, methaqualone, valium); , tranquilizers such as librium, miltown, serax, equanil, tranxine, etc.); hallucinogens (phencyclidine, lysergic acid, mescaline, peyote, psilocybin); narcotics (heroin, codeine, morphine, opium, meperidine, percocet, percodan, tasionex, fentanyl, darvon, talwin, lomotil); synthetic narcotics (fentanyl, meperidine, amphetamine, methamphetamine, and analogs of phencyclidine, e.g., ecstasy); anabolic steroids; and nicotine. Metabolic products of drugs and pharmaceutical compositions are also contemplated analytes. Analytes such as neurochemicals and other chemicals produced in the body, such as ascorbic acid, uric acid, dopamine, noradrenaline, 3-methoxytyramine (3MT), 3,4-dihydroxyphenylacetic acid (DOPAC), homovanillic acid (HVA), 5-hydroxytryptamine (5HT), and 5-hydroxyindoleacetic acid (FHIAA), may also be analyzed.

As used herein, the term "baseline" is a broad term given its ordinary and customary meaning to those of skill in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, a component of an analyte sensor signal that is not related to an estimated analyte value. In one example of a glucose sensor, the baseline is substantially composed of signal contributions due to factors other than glucose (e.g., interfering species, unreacted associated hydrogen peroxide, or other electroactive species with oxidation potentials that overlap with hydrogen peroxide). In some embodiments, the calibration may be defined by solving for the equation y = mx + b, where the value of b represents the baseline of the signal. In certain embodiments, the value of b (i.e., the baseline) may be 0 or about 0. This may be the result of, for example, a baseline subtraction electrode or a low bias potential setting. As a result, in these embodiments, the calibration may be defined by solving for the equation y = mx.

As used herein, the term "biological sample" is a broad term given its ordinary and customary meaning to those of skill in the art (and is not limited to any special or customized meaning) and refers to a sample derived from the body or tissue of a host, such as, but not limited to, blood, interstitial fluid, spinal fluid, saliva, urine, tears, sweat, or similar fluids.

As used herein, the term "calibration" is a broad term given its ordinary and customary meaning to those skilled in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, the process of determining the grade of a sensor that provides a quantitative measurement (e.g., an estimated analyte value). By way of example, calibration may be updated or recalibrated over time to account for changes associated with the sensor, such as changes in sensor sensitivity and sensor background. Additionally, calibration of a sensor may involve automatic self-calibration, i.e., calibration without the use of a reference analyte value after the point of use.

As used herein, the term "co-analyte" is a broad term given its ordinary and customary meaning to those of skill in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, a molecule required in an enzymatic reaction to react with an analyte and an enzyme to form the particular product being measured. In one embodiment of a glucose sensor, an enzyme, glucose oxidase (GOX), is provided to react with glucose and oxygen (the co-analyte) to form hydrogen peroxide.

As used herein, the term "comprising" is synonymous with "including," "containing," or "characterized by," is inclusive or open-ended, and does not exclude additional unrecited elements or methods.

As used herein, the term "computer" is a broad term given its ordinary and customary meaning to those skilled in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, a machine that can be programmed to manipulate data.

As used herein, the terms "continuous analyte sensor" and "continuous glucose sensor" are broad terms and are to be given their ordinary and customary meaning to those of skill in the art (and are not to be limited to any special or customized meaning), and refer to, but are not limited to, a device that continuously or continuously measures analyte/glucose concentration and/or calibrates the device (e.g., by continuously or continuously adjusting or determining the sensitivity and background of the sensor) over time intervals ranging from, for example, a fraction of a second to, for example, 1, 2, 5 minutes or more.

As used herein, the term "continuous glucose sensing" is a broad term given its ordinary and customary meaning to those skilled in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, periods during which monitoring of plasma glucose concentration is performed continuously or continuously, for example, at time intervals ranging from less than one second up to, for example, one, two, or five minutes or more.

As used herein, the term "counts" is a broad term given its ordinary and customary meaning to those of skill in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, a unit of measurement for a digital signal. In one example, the raw data stream measured in counts is directly related to a voltage (e.g., converted by an A/D converter) that is directly related to a current from the working electrode.

As used herein, the term "distal" is a broad term and is to be given its ordinary and customary meaning to those of skill in the art (and is not to be limited to any special or customized meaning), and refers to a space relatively far from a reference point such as an origin or attachment point, including, but not limited to, a space relatively far from a reference point such as an origin or attachment point.

As used herein, the term "domain" is a broad term and is to be given its ordinary and customary meaning to one of ordinary skill in the art (and is not to be limited to any special or customized meaning) and refers to, but is not limited to, a layer, a uniform or non-uniform gradient (e.g., an anisotropic region of a film), a region of a film that may be a functional aspect of a material, or a region of a film that is provided as part of a film.

As used herein, the term "electrical conductor" is a broad term and is to be given its ordinary and customary meaning to one of ordinary skill in the art (and is not to be limited to any special or customized meaning) and refers to, but is not limited to, a material that contains a mobile charge. When a potential difference is applied to distinct points on a conductor, the mobile charges in the conductor are forced to move and a current appears between those points according to Ohm's law.

As used herein, the term "electrical conductance" is a broad term and is given its ordinary and customary meaning to one of ordinary skill in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, the tendency of a material to behave as an electrical conductor. In some embodiments, the term refers to an amount of electrical conductance (e.g., a material property) sufficient to provide a desired function (electrical conduction).

As used herein, the terms "electrochemically reactive surface" and "electroactive surface" are broad terms and are to be given their ordinary and customary meaning to one of ordinary skill in the art (and are not to be limited to any special or customized meaning), and refer to, but are not limited to, the surface of an electrode on which an electrochemical reaction occurs. In one embodiment, the working electrode measures hydrogen peroxide (H2O2), which produces a measurable electronic current.

As used herein, the term "electrode" is a broad term and is given its ordinary and customary meaning to one of ordinary skill in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, a conductor through which electricity travels to or from something, such as a battery or an electrical device. In one embodiment, an electrode is the metal portion of a sensor (e.g., an electrochemically reactive surface) that is exposed to the extracellular environment to detect an analyte. In some embodiments, the term electrode includes a conductive wire or trace that electrically connects the electrochemically reactive surface to a connector (to connect the sensor to an electronic device) or to an electronic device.

As used herein, the term "elongated conductor" is a broad term and is to be given its ordinary and customary meaning to one of ordinary skill in the art (and is not to be limited to any special or customized meaning) and refers to, without limitation, an elongated object formed at least in part of a conductive material, including any number of coatings that may be formed on the elongated object. As an example, "elongated conductor" may mean a bare elongated conductive core (e.g., a metal wire), or an elongated conductive core coated with one, two, three, four, five, six or more layers of material, each of which may or may not be conductive.

As used herein, the term "enzyme" is a broad term and is given its ordinary and customary meaning to one of ordinary skill in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, a protein or a protein-based molecule that facilitates a chemical reaction that occurs in vivo. An enzyme may act as a catalyst for a single reaction that converts a reactant (also referred to herein as an analyte) into a specific product. In one embodiment of a glucose oxidase-based glucose sensor, an enzyme, glucose oxidase (GOX), is provided to react with glucose (the analyte) and oxygen to form hydrogen peroxide.

As used herein, the term "filtering" is a broad term given its ordinary and customary meaning to those skilled in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, the modification of a set of data to make it smoother or more continuous and to remove or mitigate outliers, for example, by performing a moving average of a raw data stream.

As used herein, the term "function" is a broad term and is to be given its ordinary and customary meaning to one of ordinary skill in the art (and is not to be limited to any special or customized meaning) and refers to, but is not limited to, an action or use for which something is suitable or designed.

As used herein, the term "GOx" is a broad term and is to be given its ordinary and customary meaning to one of ordinary skill in the art (and is not to be limited to any special or customized meaning), and refers to, but is not limited to, the enzyme glucose oxidase (e.g., GOx is an abbreviation).

As used herein, the term "host" is a broad term given its ordinary and customary meaning to those of skill in the art (and is not limited to any special or customized meaning) and refers to animals, including, but not limited to, humans.

As used herein, the term "inactive enzyme" is a broad term and is given its ordinary and customary meaning to one of skill in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, an enzyme (e.g., glucose oxidase, GOx, etc.) that has been inactivated (e.g., by denaturation of the enzyme) and has substantially no enzymatic activity. Enzymes can be inactivated using a variety of techniques known in the art (e.g., but not limited to, heating, freeze-thawing, denaturation in organic solvents, acid or base, crosslinking, genetic alteration of enzymatically important amino acids, etc.). In some embodiments, a solution containing an active enzyme can be applied to the sensor, and the applied enzyme is then inactivated by heating or treatment with an inactivating solvent.

As used herein, the terms "insulating properties," "electrical insulator," and "insulator" are broad terms given their ordinary and customary meanings to those skilled in the art (and are not limited to any special or customized meanings) and refer to, but are not limited to, the tendency of a material that does not have a mobile charge to block the movement of charge between two points. In one embodiment, an electrically insulating material may be placed between two conductive materials to block the movement of electricity between the two conductive materials. In some embodiments, these terms refer to an amount of insulating properties (e.g., of a material) sufficient to provide the required function (electrical insulation). The terms "insulator" and "non-conductive material" may be used interchangeably herein.

As used herein, the term "in vivo portion" is a broad term given its ordinary and customary meaning to those of skill in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, the portion of a device that is implanted or inserted into a host. In one embodiment, the in vivo portion of a transcutaneous sensor is the portion of the sensor that is inserted through the skin of the host and resides within the host.

As used herein, the phrase "membrane system" is a broad term and is to be given its ordinary and customary meaning to one of ordinary skill in the art (and is not to be limited to any special or customized meaning), and refers, without limitation, to a permeable or semi-permeable membrane that may contain two or more domains, is typically constructed of a material several microns thick or greater, and may be permeable to oxygen and, optionally, to glucose. In one example, the membrane system may include an immobilized glucose oxidase enzyme that allows an electrochemical reaction to occur to measure the concentration of glucose.

As used herein, the term "operably connected" is a broad term given its ordinary and customary meaning to those of skill in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, one or more components that are linked to another component in a manner that allows for the transmission of a signal between the components. For example, one or more electrodes may be used to detect the amount of analyte in a sample and convert that information into a signal, which can then be transmitted to an electronic circuit. In this instance, the electrodes are "operably linked" to the electronic circuit. These terms are broad enough to include wired and wireless connectivity.

As used herein, the term "potentiostat" is a broad term given its ordinary and customary meaning to those of skill in the art (and is not limited to any special or customized meaning) and refers to an electrical system that applies a potential at a preset value between the working and reference electrodes of a two or three electrode cell and measures the current through the working electrode. The potentiostat forces whatever current needs to flow between the working and counter electrodes to maintain the desired potential, as long as the required cell voltage and current do not exceed the compliance limits of the potentiostat.

As used herein, the terms "processor module" and "microprocessor" are broad terms that are to be given their ordinary and customary meanings to those skilled in the art (and are not to be limited to any special or customized meanings) and refer to, but are not limited to, a computer system, state machine, processor, or the like that is designed to perform arithmetic and logical operations using logic circuitry that responds to and processes the basic instructions that drive a computer.

As used herein, the term "proximal" is a broad term and is to be given its ordinary and customary meaning to one of ordinary skill in the art (and is not to be limited to any special or customized meaning), including, but not limited to, near a reference point such as an origin or attachment point.

As used herein, the terms "raw data stream" and "data stream" are broad terms given their ordinary and customary meanings to those skilled in the art (and are not limited to any special or customized meanings) and refer to, but are not limited to, an analog or digital signal directly related to an estimated analyte value measured by an analyte sensor. In one example, a raw data stream is a digital data of counts converted by an A/D converter from an analog signal (e.g., voltage or amperes) representing an estimated analyte value. The terms broadly encompass multiple time-interval data points from a substantially continuous analyte sensor, which may include individual measurements taken at time intervals ranging from a fraction of a second to, for example, 1, 2, or 5 minutes or more.

As used herein, the term "RAM" is a broad term and is to be given its ordinary and customary meaning to one of ordinary skill in the art (and is not to be limited to any special or customized meaning) and refers to, but is not limited to, a data storage device in which the order of access to different locations does not affect the speed of access. RAM is broad enough to include, for example, SRAM, which is a static random access memory that retains data bits in its memory as long as power is applied.

As used herein, the term "ROM" is a broad term given its ordinary and customary meaning to those skilled in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, read-only memory, a type of data storage device manufactured with fixed contents. ROM is broad enough to include, for example, EEPROM, which is an electrically erasable programmable read-only memory (ROM).

As used herein, the terms "reference analyte value" and "reference data" are broad terms given their ordinary and customary meanings to those of skill in the art (and are not limited to any special or customized meanings) and refer to, but are not limited to, reference data from a reference analyte monitor, such as a blood glucose meter, that includes one or more reference data points. In some embodiments, the reference glucose value is obtained, for example, from a self-monitored blood glucose (SMBG) test (e.g., from a finger or forearm blood test) or a Yellow Springs Instruments (YSI) test.

As used herein, the term "regression" is a broad term given its ordinary and customary meaning to those of skill in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, finding a line where a set of data has the least measurement (e.g., deviation) from that line. Regression can be linear, nonlinear, first order, second order, etc. One example of regression is least squares regression.

As used herein, the term "sensing area" is a broad term given its ordinary and customary meaning to those skilled in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, an area of a monitoring device responsible for the detection of a particular analyte. In one embodiment, the sensing area generally comprises a non-conductive body, at least one electrode, a reference electrode, and optionally a counter electrode, passing through and secured within the body to form an electroactive surface at one location on the body and to form an electrical connection at another location on the body, and a membrane system affixed to the body and covering the electroactive surface.

As used herein, the term "sensitivity" or "sensor sensitivity" is a broad term given its ordinary and customary meaning to those of skill in the art (and is not limited to any special or customized meaning) and refers to, but is not limited to, the amount of signal produced by a particular concentration of a measured analyte or a measured species (e.g., H2O2, etc.) associated with a measured analyte (e.g., glucose, etc.). For example, in one embodiment, the sensor has a sensitivity of about 1 to about 300 picoamps of current per mg/dL of glucose analyte.

As used herein, the terms "sensitivity profile" and "sensitivity curve" are broad terms that are to be given their ordinary and customary meanings to those of skill in the art (and are not to be limited to any special or customized meaning) and refer to, but are not limited to, a representation of the change in sensitivity over time.

As used herein, the terms "sensor analyte value" and "sensor data" are broad terms given their ordinary and customary meanings to those of skill in the art (and are not limited to any special or customized meanings) and refer to, but are not limited to, data received from a continuous analyte sensor that includes one or more spatiotemporal sensor data points.

As used herein, the terms "sensor electronics" and "electronic circuitry" are broad terms given their ordinary and customary meanings to those of skill in the art (and are not limited to any special or customized meanings) and refer to, but are not limited to, the components of a device (e.g., hardware and/or software) configured to process data. In the case of an analyte sensor, the data includes biological information obtained by the sensor regarding the concentration of an analyte in a biological fluid. U.S. Pat. Nos. 4,757,022, 5,497,772, and 4,787,398 describe suitable electronic circuitry that may be utilized with certain embodiment devices.

As used herein, the terms "sensor environment" and "sensor operating environment" are broad terms given their ordinary and customary meaning to those skilled in the art (and are not limited to any special or customized meaning) and refer to the biological environment in which a sensor operates.

As used herein, the terms "substantial" and "substantially" are broad terms having their ordinary and customary meanings given to those of ordinary skill in the art (and are not limited to any special or customized meanings) and refer primarily to, but not necessarily in whole of, what is specified.

As used herein, the term "thermal conductivity" is a broad term and is to be given its ordinary and customary meaning to one of ordinary skill in the art (and is not to be limited to any special or customized meaning), and refers to, but is not limited to, the amount of heat transferred due to a unit temperature gradient in unit time under steady-state conditions in a direction perpendicular to a surface of a unit area.

As used herein, the term "thermal coefficient" is a broad term that is to be given its ordinary and customary meaning to those skilled in the art (and is not to be limited to any special or customized meaning) and refers to, but is not limited to, the change in resistance of a material at various temperatures.

As used herein, the term "thermally conductive material" is a broad term and is to be given its ordinary and customary meaning to one of ordinary skill in the art (and is not to be limited to any special or customized meaning) and refers to, but is not limited to, a material that exhibits a high degree of thermal conductivity.

As used herein, "thermocouple" is a broad term that is to be given its ordinary and customary meaning to those skilled in the art (and is not to be limited to any special or customized meaning) and refers to, but is not limited to, a device that includes two dissimilar conductors (e.g., metal alloys, etc.) that produce a voltage proportional to the temperature difference between each end of the two conductors.

Overview Some analyte sensors measure the concentration of a substance (e.g., glucose) in the body (e.g., measuring glucose concentration in blood or interstitial fluid at a subcutaneous location). The output of the analyte sensor may be affected by temperature. The temperature of the subcutaneous area of the body where the sensor may be located may vary from person to person and may vary over time in an individual person. For example, subcutaneous temperature may be affected by body temperature changes (such as fever or cyclical fluctuations) as well as ambient temperature changes. For example, exposure to hot or cold water, warm clothing, exposure to cold weather, and sunlight may change the subcutaneous temperature of a host. When conditions or states such as these exist, temperature variations may cause inaccuracies in the estimation of glucose concentration levels. The accuracy and precision of the estimated glucose value may be improved by compensating for temperature variations within the sensing site or sensor when the sensor is worn by the host.

The performance of an analyte sensor system can be improved by compensating for these temperature effects. For example, temperature compensation can increase sensory accuracy or reduce MARD. However, temperature compensation is problematic to implement because it can be difficult to know the actual temperature at the sensing site, or how much to compensate, and the temperatures of the host's body and various system components may vary relative to each other and over time.

In some examples, temperature compensation may be applied to the sensitivity value used to convert the signal from the sensor to an estimated analyte value (e.g., a 3% change in sensitivity for every 1°C deviation from a reference temperature (e.g., 35°C)). In some examples, temperature compensation may be applied directly to the estimated glucose value. In some cases, a compensated glucose value instead of a sensor sensitivity may yield a more accurate value. For example, in addition to variations in enzyme sensitivity, other effects may affect glucose concentration levels or sensor response. Additional temperature effects may include local glucose concentration variations (as opposed to systemic glucose levels), compartment bias (differences in glucose concentrations between interstitial fluid and blood), and non-enzymatic sensor bias (e.g., electrochemical baseline signal not generated by glucose/enzyme interaction). Models may be developed to account for some or all of these additional factors, which may provide more accurate estimates of glucose concentration levels.

FIG. 1 illustrates an exemplary system 100 in which exemplary temperature compensation systems, devices, and methods may be implemented. The system 100 may include a continuous analyte sensor system 8, including sensor electronics 12 and a continuous analyte sensor 10. The system 100 may include a drug delivery pump 2 (which may be communicatively coupled to the continuous analyte sensor system, e.g., to enable closed-loop therapy), and other devices and/or sensors, such as a glucose meter 4, such as a blood glucose meter, which may be communicatively coupled to the continuous analyte sensor system 8. The continuous analyte sensor 10 may be physically coupled to the sensor electronics 12, may be releasably attachable to the sensor electronics 12, or may be integral to the sensor electronics 12 (e.g., may be non-releasably attached). The sensor electronics 12, drug delivery pump 2, and/or glucose meter 4 may be coupled to one or more devices, such as display devices 14, 16, 18, and/or 20.

In some exemplary implementations, the system 100 may include a cloud-based analyte processor 490 configured to analyze analyte data (and/or other patient related data) provided over the network 406 (e.g., via wired, wireless, or combinations thereof) from the sensor system 8 and other devices, such as display devices 14-20 associated with the host (also referred to as the subject or patient), to generate reports that provide high level information, such as statistics, regarding analytes measured over a particular time frame. A complete discussion of using a cloud-based analyte processing system may be found in U.S. Patent Application Publication No. 2013/0325352 (A1), filed March 7, 2013, entitled "Cloud-Based Processing of Analyte Data," which is incorporated herein by reference in its entirety. In some implementations, one or more steps of the temperature compensation algorithm may be implemented in the cloud.

In some exemplary implementations, the sensor electronics 12 may include electronic circuitry associated with measuring and processing data generated by the continuous analyte sensor 10. This generated continuous analyte sensor data may also include algorithms that can be used to process and calibrate the continuous analyte sensor data, although these algorithms may be provided in other ways. The sensor electronics 12 may include hardware, firmware, software, or a combination thereof for providing a measurement of an analyte value via a continuous analyte sensor, such as a continuous glucose sensor. An exemplary implementation of the sensor electronics 12 is further described below with respect to FIG. 2B. In one implementation, a temperature compensation method may be implemented by the sensor electronics 12.

The sensor electronics 12 can be coupled (e.g., wirelessly, etc.) to one or more devices, such as display devices 14, 16, 18, and/or 20, as described above. The display devices 14, 16, 18, and/or 20 can be configured to present (and/or alert) information, such as sensor information transmitted by the sensor electronics 12, for display on the display devices 14, 16, 18, and/or 20.

The display device may include a relatively small display device 14. In some exemplary implementations, the relatively small display device 14 may be part of a key fob, wrist watch, belt, necklace, pendant, jewelry, adhesive patch, pager, key fob, plastic card (e.g., credit card), identification (ID) card, etc. This small display device 14 may include a relatively small display portion (e.g., smaller than the large display device 16) and may be configured to display certain types of displayable sensor information, such as numbers and arrows, or color codes. The device 14 may be configured as a data receiving or tracking device 14 (e.g., a blood glucose meter or CGM receiver) and may include a communication device (e.g., a USB-B port or wireless communication transceiver) for uploading data to another device.

In some example implementations, the relatively large handheld display device 16 may include a handheld receiver device, a palmtop computer, and/or the like. This large display device may include a relatively larger display portion (e.g., larger than the small display device 14) and may be configured to display information such as a graphical representation of continuous sensor data, including current and historical sensor data, output by the sensor system 8. The handheld display device 16 may be, for example, a CGM controller or a pump controller.

The display device may also include a mobile device 18 (e.g., a smartphone, tablet, or other smart device). The display device may also include a computer 20 and/or any other user equipment configured to present at least information (e.g., medication delivery information, discrete self-monitoring glucose readings, heart rate monitor, caloric intake monitor, etc.).

Any of the display devices may be coupled to the network 406 via a wired or wireless (e.g., cellular, Bluetooth, Wi-Fi, MICS, ZigBee) connection and may include a processor and memory circuitry for storing and processing information. In some examples, the temperature compensation method may be implemented at least in part by one or more of the display devices.

In some exemplary implementations, the continuous analyte sensor 10 may include a sensor for detecting and/or measuring an analyte, and the continuous analyte sensor 10 may be configured to continuously detect and/or measure an analyte as a non-invasive device, a subcutaneous device, a transdermal device, and/or an intravascular device. In some exemplary implementations, the continuous analyte sensor 10 may analyze multiple intermittent blood samples, although other analytes may be used as well.

In some exemplary implementations, the continuous analyte sensor 10 may include a glucose sensor configured to measure glucose in blood or interstitial fluid using one or more measurement techniques, such as enzymatic, chemical, physical, electrochemical, spectrophotometric, polarimetric, calorimetric, iontophoretic, radiometric, immunochemical, etc. In implementations in which the continuous analyte sensor 10 includes a glucose sensor, the glucose sensor may include any device capable of measuring the concentration of glucose and may provide data, such as a data stream indicative of the concentration of glucose in a host, using a variety of techniques for measuring glucose, including invasive, minimally invasive, and non-invasive sensing techniques (e.g., fluorescence monitoring). The data stream may be sensor data (raw data and/or filtered data) and may be converted to a calibrated data stream that is used to provide glucose values to a host, such as a user, patient, or caregiver (e.g., a parent, relative, guardian, teacher, doctor, nurse, or any other individual interested in the health of the host). Additionally, the continuous analyte sensor 10 can be implanted as at least one of the following types of sensors: an implantable glucose sensor, a transcutaneous glucose sensor that is implanted in a host's blood vessel or outside the body, a subcutaneous sensor, a refillable subcutaneous sensor, and an intravascular sensor.

While the disclosure herein refers to some implementations including a continuous analyte sensor 10 comprising a glucose sensor, the continuous analyte sensor 10 may include other types of analyte sensors as well. Additionally, while some implementations may refer to the glucose sensor as an implantable glucose sensor, other types of devices capable of detecting glucose concentrations and providing an output signal representative of the glucose concentration may be used as well. Additionally, while the description herein refers to glucose as the analyte that is measured, processed, etc., other analytes may be used as well, including, for example, ketone bodies (e.g., acetone, acetoacetate and beta-hydroxybutyrate, lactate esters, etc.), glucagon, acetyl coenzyme A, triglycerides, fatty acids, intermediates in the citric acid cycle, choline, insulin, cortisol, testosterone, etc.

2A is a schematic diagram of an exemplary analyte sensor system 250, which may be, for example, system 8 shown in FIG. 1. The analyte sensor system may include an analyte sensor, such as a glucose sensor 252, one or more temperature sensors 254, a processor 251, and a memory 256. The processor may receive a glucose sensor signal indicative of a glucose concentration level from the glucose sensor 252 and a temperature sensor signal indicative of a temperature parameter (e.g., absolute or relative temperature, or a temperature gradient) from the temperature sensor 254. The sensor system 250 may also include one or more additional sensors 258, which may include, for example, a heart rate sensor, an activity sensor (e.g., an accelerometer), or a blood pressure sensor (e.g., to measure compression of the sensor against the host).

The processor 251 may determine a temperature-compensated glucose concentration level (or other estimated analyte value) based on the glucose sensor signal, the temperature sensor signal, and optionally also based on one or more signals from additional sensor 258. The processor 251 may determine a particular temperature-compensated sensitivity value (e.g., an analyte sensor sensitivity value based on temperature) or may determine a compensated estimated glucose value. The signal from the temperature sensor 254 may be used as an approximation of the temperature at the analyte sensor, or the signal from the temperature sensor 254 may be processed (e.g., using methods described in detail below) to determine an estimated analyte temperature sensor based on the signal from the temperature sensor 254. In some examples, the processor may retrieve instructions or information from memory 256 to determine a temperature-compensated glucose concentration level. For example, the processor may access a lookup table, or may apply an algorithm based on the glucose sensor signal and the temperature sensor signal, or may apply the glucose sensor signal and the temperature sensor signal to a model (e.g., using a state model or a neural network). In some examples, the processor may retrieve executable instructions from memory 256 (or a separate memory that may be operatively coupled to or integrated into the processor). In some examples, the processor may include, or be part of, an application specific integrated circuit (ASIC) that may be configured to determine a temperature compensated glucose concentration level. In various examples, any one or more of the methods described herein or illustrated in FIGS. 6-14 may be performed by the processor 251 or the temperature compensated glucose sensor, alone or in conjunction with other processors or devices, such as the device illustrated in FIG. 5.

2B shows a more detailed view of an exemplary sensor electronics 12. The sensor electronics may be part of a system of devices such as that shown in FIG. 1. The sensor electronics 12 may include sensor electronics configured to process sensor information, such as sensor data, and generate transformed sensor data and displayable sensor information, e.g., via a processor module 214. For example, the processor module 214 may convert the sensor data into one or more of the following: temperature compensated data, filtered sensor data (e.g., one or more filtered analyte values), raw sensor data, calibrated sensor data (e.g., one or more calibrated analyte values), rate of change information, trend information, acceleration/deceleration information, sensor diagnostic information, position information, alarm/alert information, calibration information as may be determined by a calibration algorithm, sensor data smoothing and/or filtering algorithms, etc.

The sensor electronics 12 may include a first temperature sensor 240. In some examples, the signal from the temperature sensor 240 may be used for temperature compensation, e.g., to compensate for the effect of temperature on the analyte sensor. In some examples, the sensor electronics 12 may include an optional second temperature sensor 242. The signals from the first temperature sensor 240 and the second temperature sensor 242 may be used to determine the heat flux or temperature gradient. In some examples, the temperature sensors 240, 242 act as backups for each other. For example, if the temperature sensor 240 fails, the sensor electronics 12 may continue to operate with the temperature signal from the temperature sensor 242. If the temperature sensor 242 fails, the sensor electronics 12 may continue to operate with the temperature signal from the temperature sensor 240.

In some embodiments, the processor module 214 may be configured to perform a substantial portion, if not all, of the data processing, including data processing for factory calibration or temperature compensation. Factory calibration may be a calibration of a continuous analyte sensor capable of achieving a high level of accuracy without (or with reduced reliance on) reference data from a reference analyte monitor (e.g., a blood glucose meter). The processor module 214 may be integrated with the sensor electronics 12 and/or may be located remotely, such as one or more of the devices 14, 16, 18, and/or 20, and/or the cloud 490. In some embodiments, the processor module 214 may include multiple smaller subcomponents or submodules. For example, the processor module 214 may include an alert module (not shown) or a prediction module (not shown), or any other suitable module that may be utilized to efficiently process data. When the processor module 214 is comprised of multiple sub-modules, the sub-modules may be located within the processor module 214, including within the sensor electronics 12 or other associated devices (e.g., 14, 16, 18, 20, and/or 490). For example, in some embodiments, the processor module 214 may be located at least in part within the cloud-based analyte processor 490 or may be located elsewhere within the network 406.

In some exemplary implementations, the processor module 214 may be configured to calibrate the sensor data, and the data storage memory 220 may store the calibrated sensor data points as transformed sensor data. Additionally, the processor module 214 may be configured to wirelessly receive calibration information from a display device, such as devices 14, 16, 18, and/or 20, in some exemplary implementations to enable calibration of the sensor data from the sensor 12. Additionally, the processor module 214 may be configured to perform additional algorithmic processing on the sensor data (e.g., calibrated and/or filtered data and/or other sensor information), and the data storage memory 220 may be configured to store the transformed sensor data and/or sensor diagnostic information associated with the algorithm. The processor module 214 may further be configured to store and use the calibration information determined from the calibration.

In some example implementations, some or all of the sensor electronics 12 may be incorporated to include an ASIC 205, which may be coupled to the user interface 222 via a wired or wireless connection. For example, the ASIC 205 may include a potentiostat 210, a telemetry module 232 for transmitting data from the sensor electronics 12 to one or more devices, such as devices 14, 16, 18, and/or 20, and/or other components for signal processing and data storage (e.g., a processor module 214 and a data storage memory 220). Although FIG. 2B illustrates an ASIC 205, other types of circuits may be used as well, including a field programmable gate array (FPGA), one or more microprocessors configured to provide some, if not all, of the processing performed by the sensor electronics 12, analog circuits, digital circuits, or combinations thereof. In addition, ASIC 205 may include only a subset (one or more) of the devices, and any of devices 210, 214, 216, 218, 220, 232, 240, 242 may be included within the ASIC, or may be provided as individual components, or may be integrated as a separate ASIC (e.g., as a second ASIC or a third ASIC).

In the example shown in FIG. 2B, via a first input port for sensor data, the potentiostat 210 may be coupled to a continuous analyte sensor 10, such as a glucose sensor, to generate sensor data from the analyte. The potentiostat 210 may also provide a voltage to an analyte sensor, such as the continuous analyte sensor 10 (shown in FIG. 5) or the sensors shown in FIG. 2C, 3, 4, 5A, or 5B, via a data line 212 to bias the sensor for measurement of a value (e.g., current, etc.) indicative of an analyte concentration in a host (also referred to as the analog portion of the sensor). The potentiostat 210 may have one or more channels, depending on the number of working electrodes in the continuous analyte sensor 10.

In some exemplary implementations, the potentiostat 210 may include a resistor that converts the current value from the sensor 10 into a voltage value, and in some exemplary implementations, a current-to-frequency converter (not shown) may also be configured to continuously integrate the measured current value from the sensor 10 using, for example, a charge counting device. In some exemplary implementations, an analog-to-digital converter (not shown) may digitize the analog signal from the sensor 10 into so-called "counts" to enable processing by the processor module 214. The resulting counts may be directly related to the current measured by the potentiostat 210, which may in turn be directly related to an analyte level, such as a glucose level, in the host.

The telemetry module 232 may be operatively connected to the processor module 214 and may provide hardware, firmware, and/or software to enable wireless communication between the sensor electronics 12 and one or more other devices, such as a display device, a processor, a network access device, etc. Various wireless radio techniques that may be implemented in the telemetry module 232 include Bluetooth, Bluetooth Low-Energy, ANT, ANT+, ZigBee, IEEE 802.11, IEEE 802.16, cellular radio access techniques, radio frequency (RF), infrared (IR), paging network communication, magnetic induction, satellite data communication, spread spectrum communication, frequency hopping communication, short-range wireless communication, etc. In some example implementations, the telemetry module 232 may include a Bluetooth chip, although Bluetooth technology may also be implemented in a combination of the telemetry module 232 and the processor module 214.

The processor module 214 may control the processing performed by the sensor electronics 12. For example, the processor module 214 may be configured to process data (e.g., counts) from the sensor, filter the data, calibrate the data, perform fail-safe checks, etc.

In some exemplary implementations, the processor module 214 may include a digital filter, such as an infinite impulse response (IIR) filter or a finite impulse response (FIR) filter. The digital filter may smooth the raw data stream received from the sensor 10. Generally, the digital filter is programmed to filter the sampled data at predetermined time intervals (also referred to as a sample rate). In some exemplary implementations, the potentiostat 210 is configured to measure an analyte (e.g., glucose, etc.) at discrete time intervals, and these time intervals determine the sampling rate of the digital filter. In some exemplary implementations, the potentiostat 210 may be configured to continuously measure an analyte using, for example, a current-to-frequency converter. In these current-to-frequency converter implementations, the processor module 214 may be programmed to request a digital value from an integrator of the current-to-frequency converter at predetermined time intervals (acquisition times). These digital values acquired by the processor module 214 from the integrator may be averaged over the acquisition time due to the continuous nature of the current measurements. The acquisition time may therefore be determined by the sampling rate of the digital filter.

The processor module 214 may further include a data generator (not shown) configured to generate data packages for transmission to devices such as the display devices 14, 16, 18, and/or 20. Additionally, the processor module 214 may generate data packets for transmission to these external sources via the telemetry module 232. In some example implementations, the data packages may be customizable for each display device, as previously described, and/or may include any available data, such as temperature or temperature-related information, temperature compensation data, accelerometer data, motion data, location data, timestamps, displayable sensor information, converted sensor data, identification codes for the sensors and/or sensor electronics 12, raw data, filtered data, calibrated data, rate of change information, trend information, error detection or correction, temperature information, or any combination thereof.

The processor module 214 may also include program memory 216 and other memory 218. The processor module 214 may be coupled to a communications interface, such as a communications port 238, and a power source, such as a battery 234. Additionally, the battery 234 may be further coupled to a battery charger and/or regulator 236 to power the sensor electronics 12 and/or charge the battery 234.

The program memory 216 may be implemented as a semi-static memory for storing data such as an identifier (e.g., a sensor identifier (ID)) of the coupled sensor 10 and for storing code (also referred to as program code) for configuring the ASIC 205 to perform one or more of the operations/functions described herein. For example, the program code may configure the processor module 214 to process and filter data streams or counts, perform calibration methods, perform fail-safe checks, etc.

The memory 218 may also be used to store information. For example, the processor module 214 including the memory 218 may be used as a cache memory for the system, and temporary storage may be provided for recent sensor data received from the sensors. In some example implementations, the memory may include memory storage components such as read only memory (ROM), random access memory (RAM), dynamic RAM, static RAM, non-static RAM, easily erasable programmable read only memory (EEPROM), rewritable ROM, flash memory, etc.

The data storage memory 220 may be coupled to the processor module 214 and may be configured to store various sensor information. In some exemplary implementations, the data storage memory 220 stores one or more days of continuous analyte sensor data. For example, the data storage memory may store 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 20, and/or 30 days (or more) of continuous analyte sensor data received from the sensor 10. The stored sensor information may include one or more of the following: temperature information or temperature related information, temperature compensated data, timestamps, raw sensor data (one or more raw analyte values), calibrated data, filtered data, converted sensor data, and/or any other displayable sensor information, calibration information (e.g., reference BG values and/or previous calibration information such as obtained from a factory calibration), sensor diagnostic information, temperature information, etc.

The user interface 222 may include various interfaces, such as one or more buttons 224, a liquid crystal display (LCD) or organic light emitting diode (OLED) display 226, a vibrator 228, an audio transducer (e.g., a speaker) 230, a backlight (not shown), etc. Such components comprising the user interface 222 may provide controls for interacting with a user (e.g., a host). The one or more buttons 224 may enable, for example, toggles, menu selections, option selections, status selections, yes/no responses to on-screen questions, an "off" function (e.g., for an alarm), an "acknowledge" function (e.g., for an alarm), a reset, etc. The display 226 may provide, for example, visual data output to the user. The audio transducer 230 (e.g., a speaker) may provide audible signals in response to the triggering of certain alerts, such as current and/or predicted hyperglycemic and hypoglycemic conditions. In some example implementations, the audible signal may be distinguished by tone, volume, duty cycle, pattern, duration, etc. In some example implementations, the audible signal may be configured to be muted (e.g., acknowledged or turned off) by pressing one or more buttons 224 on the sensor electronics 12 and/or signaling the sensor electronics 12 using a button or selection on a display device (e.g., a key fob, cell phone, etc.).

Although audio and vibration alarms are described with respect to FIG. 2B, other alarm mechanisms may be used as well. For example, in some example implementations, a tactile alarm is provided that includes a pecking mechanism configured to "poke" or physically contact the patient in response to one or more alarm conditions or states.

The battery 234 may be operatively connected to the processor module 214 (and possibly other components of the sensor electronics 12) and may provide the necessary power to the sensor electronics 12. In some exemplary implementations, the battery may be a lithium manganese dioxide battery, but any suitable size and power battery (e.g., AAA battery, nickel-cadmium battery, zinc-carbon battery, alkaline battery, lithium battery, nickel-metal hydride battery, lithium-ion battery, zinc-air battery, zinc-mercury oxide battery, silver-zinc battery, or sealed battery) may be used. In some exemplary implementations, the battery may be rechargeable. In some exemplary implementations, multiple batteries may be used to power the system. In still other implementations, the receiver may be powered transcutaneously, for example, via inductive coupling.

The battery charger and/or regulator 236 may be configured to receive energy from an internal and/or external charger. In some exemplary implementations, the battery regulator (or balancer) 236 regulates the recharging process by diverting excess charging current to allow all cells or batteries in the sensor electronics 12 to be fully charged without overcharging other cells or batteries. In some exemplary implementations, the battery 234 (or batteries) may be configured to be charged via an inductive and/or wireless charging pad, although any other charging and/or power mechanism may be used.

One or more communication ports 238, also referred to as external connectors, may be provided to enable communication with other devices, for example, a PC communication (com) port may be provided to enable communication with systems separate from or integral to the sensor electronics 12. The communication ports may include, for example, a serial (e.g., Universal Serial Bus or "USB") communication port, and may enable communication with another computer system (e.g., a PC, a personal digital assistant or "PDA", a server, etc.). In some example implementations, the sensor electronics 12 may be capable of transmitting historical data to a PC or other computing device for retrospective analysis by the patient and/or HCP. As another example of data transmission, factory information may be transmitted from a sensor to the algorithm, or from a cloud data source to the algorithm.

The one or more communication ports 238 may further include a second input port through which calibration data may be received and an output port that may be used to transmit the calibrated data or the data to be calibrated to a receiver or mobile device. It will be appreciated that while the ports may be physically separated, in alternative implementations, a single communication port may provide the functionality of both the second input port and the output port.

In some continuous analyte sensor systems, the on-skin portion of the sensor electronics may be simplified to minimize the complexity and/or size of the on-skin electronics, for example, providing only raw, calibrated, and/or filtered data to a display device configured to perform calibration and other algorithms required to display the sensor data. However, the sensor electronics 12 (e.g., via the processor module 214) may be implemented to perform prospective algorithms used to generate transformed sensor data and/or displayable sensor information, including, for example, algorithms to evaluate clinical acceptability of the reference and/or sensor data, evaluate the calibration data for best calibration based on inclusion criteria, evaluate the quality of the calibration, compare estimated analyte values to time-corresponding measured analyte values, analyze variability in estimated analyte values, evaluate stability of the sensor and/or sensor data, detect signal artifacts (noise), replace signal artifacts, determine rate of change and/or trends in the sensor data, perform dynamic and intelligent analyte value estimation, perform diagnostics on the sensor and/or sensor data, set operating modes, evaluate the data for anomalies, etc.

Although separate data storage and program memory are shown in FIG. 2B, various configurations may be used as well. For example, one or more memories may be used to provide storage space to support data processing and storage requirements in the sensor electronics 12.

In a preferred embodiment, the analyte sensor may be an implantable glucose sensor, such as described with reference to U.S. Pat. No. 6,001,067 and U.S. Pat. Appl. Pub. No. 2005/0027463 (A1). In another preferred embodiment, the analyte sensor may be a transdermal glucose sensor, such as described with reference to U.S. Pat. Appl. Pub. No. 2006/0020187 (A1). In yet other embodiments, the sensor may be configured to be implanted within or outside the host's blood vessels, such as described in U.S. Pat. Appl. Pub. No. 2007/0027385 (A1), U.S. Pat. Appl. Pub. No. 2008/0119703 (A1) (now abandoned), U.S. Pat. Appl. Pub. No. 2008/0108942 (A1) (now abandoned), and U.S. Pat. No. 7,828,728. In an alternative embodiment, the continuous glucose sensor may include a transdermal sensor, such as described in U.S. Pat. No. 6,565,509 to Say et al. In another alternative embodiment, the continuous glucose sensor may include a subcutaneous sensor, such as described with reference to U.S. Patent No. 6,579,690 to Bonnecaze et al., or U.S. Patent No. 6,484,046 to Say et al. In another alternative embodiment, the continuous glucose sensor may include a refillable subcutaneous sensor, such as described with reference to U.S. Patent No. 6,512,939 to Colvin et al. In another alternative embodiment, the continuous glucose sensor may include an intravascular sensor, such as described with reference to U.S. Patent No. 6,477,395 to Schulman et al. In another alternative embodiment, the continuous glucose sensor may include an intravascular sensor, such as described with reference to U.S. Patent No. 6,424,847 to Mastrototaro et al.

FIG. 2C is a schematic diagram of an exemplary analyte sensor system 8 showing an analyte sensor 10 inserted through the epidermis 260, the dermis 262 and into the subcutaneous layer 264 such that the distal end 280 of the analyte sensor 10 is in the subcutaneous layer. In a human host, the epidermal layer 260 may typically be about 0.01 cm thick, the dermal layer 262 may typically be about 0.2 cm thick, and the subcutaneous layer may be substantially thicker, e.g., 1 cm to 1.5 cm. The working portion 282 (e.g., working electrode) of the analyte sensor 10 may be at or near the distal end 280 of the analyte sensor at a depth of about 0.5 cm. The working portion 282 may include, for example, a coating on the conductive portion 286 (e.g., conductive core). The working portion 282 may be configured to generate, for example, a voltage proportional to a glucose concentration (e.g., the working portion may be part of a glucose sensor available from Dexcom, Inc.). In some examples, the temperature sensor 284 may be provided at or near the distal end 280 of the analyte sensor. The temperature sensor 284 may be used to compensate for temperature variations using one or more of the various techniques described below. In addition, empirical measurements (discussed below and shown in FIG. 21 ) indicate that the conductance of the analyte sensor may be strongly dependent on temperature. In some examples, this relationship between conductance and temperature may be used to estimate the subcutaneous temperature, which may be used in a temperature compensation model or other methods. In other examples, the relationship between conductance and temperature may be applied directly (e.g., without using an estimated temperature) to compensate for temperature variations.

The analyte sensor can be coupled to a base 274, which can be coupled to a housing 266. The housing can house some or all of the components shown in FIG. 2A or the sensor electronics 12 shown in FIG. 2B.

In some examples, the housing may include a heat shield 272 on the top surface (and optionally additionally on one or more side surfaces) to reflect heat away from the housing, thereby, for example, reducing the effect of sunlight on the sensor 10.

In some examples, the sensor electronics 12 may include a first temperature sensor 268 near the bottom of the housing 266 and a second temperature sensor 270 near the top of the housing. A circuit such as the processor 251 or processor module 214 may be configured to determine a compensated glucose concentration level based, at least in part, on the glucose signal, the first temperature signal, and the second temperature signal.

In some examples, the temperature gradient or heat flux can be determined (e.g., by the processor 251 or the processor module 214) from the signals received from the first temperature sensor 268 and the second temperature sensor 270. For example, if the housing is exposed to sunlight, the signal from the second temperature sensor 270 may indicate a higher temperature than the signal from the first temperature sensor 268. This information can be used, for example, to estimate the temperature at the analyte sensor 10 or can be used in a temperature compensation algorithm or model. In another example, the sensor can be exposed to a cold temperature, in which case the second temperature sensor 270 may indicate a lower temperature than the first temperature sensor. In another example, the system can be immersed in cold water, in which case the first temperature sensor 268 and the second temperature sensor may initially indicate a gradient but quickly transition to approximately equal temperature values. This information may be used directly in temperature compensation based on the relationship between one or more of the temperature sensors 268, 270 and the temperature at the analyte sensor 10, or the temperature information may be used indirectly as an indication of the analyte sensor or host environment (e.g., immersed in hot or cold water, exposed to cold air, exposed to the sun) from which temperature or temperature compensation information may be inferred or which may be applied to a model.

FIG. 2D is a schematic diagram of another exemplary configuration of analyte sensor system 8 engaged with host tissue. In the example of FIG. 2D, temperature sensor 281 is positioned on base 274 in contact with epidermis 260 of the host's skin. For example, temperature sensor 281 may be incorporated into an adhesive pad for adhering base 274 to the host's skin.

FIG. 3 is a schematic diagram of an exemplary distal portion 11 of an analyte sensor 10 that may include an analyte sensor region 302 configured to generate a sensor signal indicative of a glucose concentration level of a host substance (e.g., interstation fluid). The signal may be conducted to one or more elongated members 304, 306, which may be wires (e.g., platinum or tantalum or alloys thereof). The sensor signal may be communicated to sensor electronics for processing. The analyte sensor 10 may also include a temperature sensor 308 that may be at or near the analyte sensor region 302. In one example, the temperature sensor 308 may be a thermocouple that may produce a voltage proportional to a temperature difference between, for example, a junction 310 of the conductors 304, 306 and a second junction (not shown) that may be at a proximal end of the conductors (e.g., outside the host). To form a working thermocouple, the conductors 304, 306 may be formed of different materials. For example, one of the conductors 304, 306 can be platinum and the other of the conductors 304, 306 can be tantalum. The signal generated by the thermocouple can be communicated to the sensor electronics for processing (i.e., for use in compensating the glucose sensor value for temperature).

In another example, the temperature sensor 308 can be a thermistor. The resistance of the thermistor can be measured using the conductors 304, 306 and communicated to the sensor electronics for processing.

In one example, a sequential method may be used to measure glucose concentration levels and temperature using a pair of conductors (e.g., a platinum conductor and a tantalum conductor as described above, which may be 304, 306 in FIG. 3). For example, an estimated analyte value may be measured by applying a voltage (e.g., 0.6 volts) across the conductors, and a temperature measurement may then be obtained by measuring the open circuit potential across the conductors, or by applying a low voltage input across the conductors and measuring the current (e.g., to determine the resistance of the thermistor and thereby determine a temperature parameter).

In another example, the temperature sensor may be positioned at the proximal end of the sensor wire, and the sensor wire may have a high thermal conductivity such that the temperature measurement at the proximal end approximates the temperature near the analyte sensor (i.e., the distal end). In various examples, such approximate temperature measurements may be used for temperature compensation.

FIG. 4 is a schematic diagram of an exemplary proximal portion 401 and electrical contact portion 402 of an analyte sensor 10, which may be part of, for example, the sensor electronics or a transmitter (such as a transmitter manufactured by Dexcom and configured to mate with a base portion including a subcutaneous glucose sensor). The proximal portion 402 of the analyte sensor may include a first conductor 404 and a second conductor 406, which may have a distal end (not shown) coupled to an analyte sensor (e.g., a glucose sensor). The electrical contact portion 404 may include a first contact 412 configured to contact the first conductor 404 and a second contact configured to contact the second conductor 406. The proximal portion may also include a thermistor 408 and a third conductor 411 coupled to the thermistor and configured to mate with a third contact 414 on the electrical contact portion. The temperature sensitive resistance of the thermistor can be used to compensate for the effect of temperature on the glucose sensor.

Figure 5A is a diagram of a configuration similar to that of Figure 4, but with the thermistor of Figure 4 replaced with a temperature sensitive coating 508. Figure 5B is a close-up view of the temperature sensitive coating 508 that may be on the conductor 406. A conductive element 510 may be configured to couple with the coating or may be connected to the coating and may be configured to couple with a third contact 414, such that the resistance of the coating may be measured using applying a voltage or driving a current across the contacts 412, 414.

The system may use a learned or defined relationship between an input (e.g., a temperature sensor signal, or one or more other sensor signals) and an analyte level to compensate for the effect of temperature on an analyte sensor (e.g., a glucose sensor) and provide an estimate (e.g., an estimated glucose value) that is less affected by temperature variations. The relationship may be defined, for example, by a theoretical model, determined from bench data, clinical trial data, or a combination thereof.

Various techniques and models or algorithms can be applied or combined to compensate for temperature signal variations caused by temperature changes. For example, the system may compensate for long-term trends or averages, or may compensate for short-term (e.g., real-time) changes, or a combination thereof.

In some examples, a linear relationship between temperature and the glucose sensor signal may be determined and used to approximate the relationship between temperature and the glucose sensor signal and compensate for the effects of temperature, for example, as given by the following equation [1]:

In some examples, the sensitivity M(t) of the sensor to the analyte (glucose) concentration may be compensated for the effects of temperature by determining a temperature compensated sensitivity M(t)comp based on a programmed (e.g., factory calibrated) sensitivity M(t)pro and a percent sensitivity change (Z) per degree Celsius. The temperature difference (ΔT) may be determined as the difference between a sensed or determined subcutaneous temperature, _Tsubcu (t), at time t and a reference temperature, _{Tsubcu,reference,} for example, as given by the following equation [2]:

The reference temperature, T _{subcu,reference} , may be, for example, an average or predetermined subcutaneous temperature value. The compensated analyte sensitivity may be determined by solving an equation such as the following equation [3]:

The value of Z may be determined from bench testing for a particular sensor configuration. In some examples, Z may be modeled as a function of time, Z(t), where t may be measured from the start of the sensor session. Equation [3] is solved for M(t) _comp to obtain the compensated analyte sensitivity, for example, as given by equation [4].

The temperature compensated analyte sensitivity, M(t) _comp , may be used to convert the raw analyte sensor data to an estimated glucose value, for example, using equation [5] below:

In some examples, the offset may be determined for a particular analyte sensor design configuration, as is routinely done with existing commercially available sensors. In other examples, multiple blood glucose measurements (or biological samples, in the case of other analytes) may be obtained (e.g., via a user interface) and used to determine the offset for a particular sensor.

In some examples, the temperature compared to the reference value is a long-term average temperature. In some examples, this is to account for body temperature differences between hosts (patients). In other examples, real-time compensation for temperature may correct for temperature-based sensor fluctuations that may be caused, for example, by exposure to hot water (e.g., showering), cold water (e.g., swimming), air conditioning, sunlight, thermal fluctuations during sleep (e.g., trapped heat from a warm blanket), or other hot or cold environments. Some examples may combine long-term and real-time compensation methods.

In some examples where the temperature sensor is not subcutaneous, a delay parameter may also be used (along with a linear model, or a more complex model as described below) to compensate for the delay between the detection of a temperature change at the temperature sensor and the actual temperature change at the analyte sensor. Various exemplary methods for determining subcutaneous temperature based on a signal from a non-subcutaneous sensor are provided below.

In some systems, devices, or methods, subcutaneous or other in vivo temperatures may be determined (e.g., estimated) using a temperature signal from a non-subcutaneous temperature sensor, such as a temperature sensor in the sensor electronics of an external device (e.g., a transmitter) that may be coupled to a subcutaneous analyte (e.g., glucose) sensor. One or more of a variety of methods may be used to determine the subcutaneous temperature from the temperature signal received from the non-subcutaneous temperature sensor.

In some examples, a linear relationship between the non-subcutaneous and subcutaneous temperature values may be used to approximate the subcutaneous temperature. In some examples, a delay parameter may also be used (along with a linear model, or a more complex model as described below) to compensate for the delay between the detection of a temperature change at the temperature sensor and the actual temperature change at the analyte sensor. In some examples, a non-linear relationship (e.g., a quadratic equation or a higher order polynomial or other relationship) may be determined and used to compensate the temperature, optionally including a delay parameter. In some examples, the relationship may be determined by solving a differential equation (e.g., a heat transfer relationship) to determine the temperature compensation. For example, the sensor system may solve a differential equation each time an analyte value is needed (e.g., every 5 minutes or every 15 minutes) to provide a temperature compensated analyte value. In another example, a filter based on a differential equation or a predetermined relationship may be applied to compensate the temperature.

In some examples, the subcutaneous or other in vivo temperature may be determined from the non-subcutaneous temperature using a linear model. The linear model may be developed, for example, from a biothermal model (e.g., Penne's bioheat equation), known host tissue parameters (e.g., typical heat transfer parameters for human skin and subcutaneous tissue), and sensor electronics (e.g., transmitter) parameters, which may be determined, for example, using bench testing. The tissue parameters may include, for example, thermal conductivity or heat flux across the tissue.

Subcutaneous temperature (Tsubcu) can be determined from the measured non-subcutaneous temperature (Texternal) using a linear equation such as the following equation [6]:

In the example of equation [6], the gain/slope (a) and offset (b) may be determined, for example, using empirical data, theoretical data, or model data, or a combination thereof.

In some examples, when the analyte temperature sensitivity is known, the gain and offset for the above formula may be determined or updated based on the analyte calibration value (e.g., blood glucose value). In other words, when the confidence in the glucose sensitivity is high, the temperature may be estimated based on the blood glucose value from the finger prick and the signal received from the glucose sensor. The system may calculate the actual analyte sensitivity from using the input glucose value, then determine the subcutaneous temperature from the actual analyte sensitivity, and then determine the relationship (e.g., gain and offset values) between the subcutaneous temperature and the signal from the non-subcutaneous temperature sensor. The system may determine or receive a temperature sensor value during calibration (e.g., from a temperature sensor in the external sensor electronics) to ensure that an updated temperature sensor signal is used in determining the sensitivity, gain, and offset. In some examples, rather than using new gains and offsets, a weighted average or probabilistic model may be used to ensure that the gains and offsets are not overly influenced by independent factors that may change sensitivity, such as a period of inaccuracy after initial placement of the sensor (e.g., the "dip and recovery" phenomenon where the sensor signal produces a low sensor signal (dip) during an initial "warm-up" period, followed by a more accurate (recovered) reading following warm-up).

In some examples, the system may account for the delay between the time a temperature change is recorded at a non-subcutaneous temperature sensor and the time the temperature change actually occurs at a subcutaneous analyte (glucose) sensor. If the analyte sensing system includes a subcutaneous temperature sensor, a direct subcutaneous temperature measurement may be used for temperature compensation, but if the system relies on a non-subcutaneous (e.g., external) temperature sensor, the accuracy of the temperature compensation method may be improved by accounting for delayed temperature changes in the subcutaneous glucose sensor.

For example, the linear model described above assumes that the subcutaneous temperature matches the external temperature (e.g., sensor electronics or transmitter temperature), but that skin tissue warms and cools much slower than the transmitter, and therefore there is a delay between the time the external sensor records a temperature change and the time the change occurs at the subcutaneous location. For example, if a person walks from a cold air-conditioned room to a warmer location, the external sensor records a temperature change quickly, but it takes much longer for the subcutaneous temperature to increase. In another example, when the host and sensor are immersed in cold water (e.g., a pool, ocean, or lake that has a temperature lower than the ambient temperature), a drop in temperature is first detected at the sensor in the external sensor electronics (e.g., in the CGM transmitter), and after a time, the temperature at the subcutaneous sensor drops due to heat loss through the sensor or host tissue. The accuracy of the subcutaneous temperature estimate can be improved by incorporating a delay to reflect this reality.

In some examples, the delay may account for a delay in recording a temperature change at the non-subcutaneous temperature sensor based on other temperature information or a model or estimate of the time delay for the non-subcutaneous sensor to record a temperature change. For example, when an environmental temperature change occurs, the period it takes for the non-subcutaneous temperature sensor to record the change may be relatively short (e.g., 1 minute), especially if the sensor is embedded within a sensor electronics housing through which heat must be conducted to record the temperature change. After some time (e.g., 6 minutes), a subcutaneous temperature change may be observed) and the net delay is the difference between the two readings (e.g., 5 minutes).

In some examples, a fixed delay may be used. For example, the temperature compensation method may assume a delay period d and compensate for the effects of temperature using temperatures from a previous period based on the assumed delay (e.g., using the temperature at time t-d). In other examples, compensation may be made using both the temperature at the current time (t) and temperatures from a previous period based on the assumed delay (e.g., using the temperature at time t-d). In yet other examples, compensation may be made using multiple temperature measurements from different periods associated with the assumed delay (e.g., using both the temperature at t-d ₁ and the temperature at t-d ₂ ). In some examples, the delay may be, for example, 30 seconds to 4 minutes (e.g., 1 minute), 1 minute to 10 minutes (e.g., 5 minutes), 5 minutes to 15 minutes (e.g., 10 minutes), or 20 minutes to 1 hour (e.g., 30 minutes). In some examples, the delay may be determined based on information known about the host, such as average body temperature or body mass index.

In some examples, a variable delay period may be used. In some examples, the variable delay period may be based, for example, at least in part, on the fluctuation between the detected temperature and a baseline. In another example, the delay may be based, at least in part, on the difference or rate of change between the detected temperature and a previously detected temperature (e.g., a longer delay may be used when a larger temperature change is observed because it takes longer for the heat transfer process to complete bringing the subcutaneous temperature to a steady state). In some examples, the delay may be implemented only when a condition is met, for example, when a sudden temperature change that exceeds a threshold occurs (e.g., a change of more than 5° C. or 10° C.).

In some examples, the variable delay may be based on a temperature gradient, e.g., the difference between the sensed temperature and the determined subcutaneous temperature. In some examples, the variable delay may be based on a heat transfer equation or model that may take into account, for example, a temperature gradient (e.g., between the ambient temperature and the subcutaneous temperature) and one or more heat transfer rates, and may optionally also take into account biological processes (e.g., heat transfer via blood flow).

Delays may be calculated or used in a variety of other exemplary ways (e.g., partial differential equation models, polynomial models, state models, time series models, models with subgroups or conditions).

FIG. 6 is a flow chart diagram of an example method 600 for determining a temperature compensated glucose concentration level using a delay parameter. The method 600 may include, at operation 602, receiving a temperature signal indicative of a temperature parameter of an external component. The temperature parameter may be, for example, a temperature, a temperature change, or a temperature offset. Detecting the temperature signal may include, for example, measuring a temperature parameter of a component of a wearable glucose sensor. The method 600 may include, at operation 604, receiving a glucose signal indicative of an in vivo glucose concentration level. Receiving the glucose signal may include, for example, receiving a glucose signal from a wearable glucose sensor.

The method 600 may include, at operation 606, determining a compensated glucose concentration level based on the glucose signal, the temperature signal, and the delay parameter. In some examples, a temperature compensated sensor sensitivity value may be determined based on the temperature signal and the delay parameter, and an estimated glucose value may be determined using the sensor sensitivity value and the glucose signal. In some examples, a model or neural network may be used to determine the estimated glucose value based (at least in part) on the glucose signal, the temperature, and the delay parameter.

In various examples, the delay parameter may be constant or may be variable based on temperature or information about the host or other factors, as described above. In some examples, the temperature parameter may be detected at a first time and the glucose concentration level may be detected at a second time after the first time. The delay parameter may include a delay period between the first time and the second time that accounts for a delay between a first temperature change in the external component and a second temperature change proximate the glucose sensor. In some examples, determining the compensated glucose concentration level may include executing instructions on the processor to receive a glucose signal and a temperature signal and determine the compensated glucose concentration level using the glucose signal, the temperature signal, and the delay parameter. The method may also include storing a value corresponding to the temperature parameter in a memory circuit and retrieving the stored value from the memory circuit for use in determining the compensated glucose concentration level. In some examples, the temperature-compensated glucose concentration level, the estimated subcutaneous temperature, or the delay parameter (or any combination thereof) may be determined using a linear model (e.g., a linear equation), a nonlinear model, a partial differential equation model, a time series model, a linear or nonlinear model with subgroups, or any other technique described herein.

The method may further include, at operation 608, adjusting the delay period based on a temperature change rate or temperature gradient (or other factors or techniques as described above) or based on a detected condition or state. In some examples, the detected condition or state may include a sudden change in temperature, location, or a motion condition or state or session (e.g., using an accelerometer).

Optionally, the method may further include, at operation 610, delivering a therapy based at least in part on the compensated glucose concentration level.

In some examples, the subcutaneous temperature may be determined from the non-subcutaneous temperature sensor signal using a partial differential equation (PDE) model. A PDE approach to temperature compensation may make the system more accurate, for example, by taking into account the fact that the rate of change of temperature in the external electronics (e.g., CGM transmitter) is higher than the rate of change of temperature in the subcutaneous tissue or fluid. The temperature of the subcutaneous tissue and fluid may change more slowly in part because the body acts as a heat sink. The use of a PDE model may be particularly advantageous in cases of rapid temperature changes.

In one example, the sensor electronics, subcutaneous sensor, and skin layers can be treated as a multi-layer model. The sensor and skin layers (epidermis 260, dermis 262, and subcutaneous tissue 264) are shown in FIG. 2C. In one example, the multi-layer structure can be considered as a one-dimensional (1D) system, where the 1D space is the depth relative to the skin surface. The distribution of temperature in space and time can be described by a heat equation, such as Penne's bioheat equation, described by the following equation [7]:

The variables and parameters from equation [7] are described by Table 1 given below.

The thermal conductivity of each layer in the 1D model may be determined or estimated. For example, the thermal conductivity of each skin layer may be determined by empirical or theoretical methods. The thermal conductivity of the sensor electronics (including the battery and epoxy adhesive) may also be determined. Exemplary values of the thermal conductivity of the different layers of the model of the sensor electronics, subcutaneous sensor, and skin layer are given by Table 2 below.

The outer boundary condition (BC) of the example PDE in equation [7] can be set to be the time-varying temperature as measured by a non-subcutaneous temperature sensor, and the inner BC can be set to be the constant core body temperature.

Based on these assumptions, equation [7] or a similar PDE may be solved so that the temperature at the sensor (e.g., at the working electrode of the sensor) may be estimated. In some examples, equation [7] or a similar PDE may be solved each time a temperature value is needed. For some values, a lookup table may be developed by solving equation [7] or a similar PDE over a range of reasonable values. The lookup table may be referenced to determine the approximate subcutaneous temperature.

In some examples, a linear correlation between the temperature at the subcutaneous sensor and the temperature at the external sensor can be determined from the PDE model. In some examples, the PDE model can be used to perform time-spatial filtering to capture transient processes of temperature change and time lag.

The estimated temperature at the subcutaneous sensor may be used to correct for sensitivity changes at the subcutaneous sensor. In one example, the temperature at the electrochemically reactive surface of an analyte sensor (e.g., a glucose sensor) may be estimated and used to determine the estimated sensitivity of the electrochemical sensor at the estimated temperature.

In some examples, a time series model may be used to estimate subcutaneous temperature using a signal from a non-subcutaneous temperature sensor or to compensate for the effect of temperature on analyte sensor sensitivity. In some examples, the temperature compensated sensitivity may be determined directly, i.e., without estimating subcutaneous temperature.

In one example, a fourth-order polynomial may be used as a model. For example, the model given by the following equation [8] may be used:

The variables and parameters from equation [8] are described by Table 3 below.

The model parameters may be determined from an empirical data set using, for example, curve fitting or optimization techniques. After the model parameters are determined, the model may be used to compensate for temperature variations. For example, the compensated sensitivity may be determined using the following equation [9]:

In some examples, the model parameters may be updated when calibration entries (e.g., based on blood glucose meter data) are available. For example, the time series model may be converted into a recursive version of the model, so that the model can be updated in real-time as fingerstick measurements are available. The values of the constants may be determined based on population data, patient-specific data. The values may be, for example, as follows: _p1 : .about.0.0004334, _p2 : 0.04955, _p3 : .about.2.035, _p4 : 36.7, p5: .about.259.7.

Although a fourth-order polynomial is provided as an example, third-order or fifth-order or higher polynomials may be used. Higher order polynomials may provide greater accuracy in compensation, but may require more time, input data, or processing power to determine and update the model parameters.

An algorithm or model may be used to determine a temperature-compensated analyte sensor value (e.g., glucose concentration level). In some examples, neural networks, state models (e.g., hidden Markov), probabilistic models, or other models may be used to develop the temperature compensation model. A model may be trained for a particular subject (e.g., a patient) based, for example, on data from the subject, and the model may be used to determine a compensated estimated glucose value. In some examples, a model may be trained from data from a population of patients (e.g., clinical trial data), and the model may be used for the population of patients. In some examples, the same model may be used for most or all patients (subject to exclusion criteria). In some examples, a patient may be matched to a model developed from a population of similar patients (e.g., based on average body temperature, age, sex, BMI, or other factors). Inputs to the model may include temperature measurements, time, sensor sensitivity, estimated glucose value, insulin sensitivity, accelerometer data (e.g., to detect activity or posture), heart rate, respiration rate, meal state, meal amount, or type, insulin on-board or insulin delivery amount or pattern, body mass index (BMI), or other factors. Output from the model may include sensor sensitivity, local glucose level, compartment bias value, non-enzymatic bias level (any of which may be combined to determine a glucose concentration level), or the model may output a compensated glucose/analyte concentration value. Model-based approaches may be particularly effective because the effects of various temperatures (e.g., sensor sensitivity, local glucose level, compartment bias value, non-enzymatic bias level) may be linear, non-linear, or dynamic (e.g., dependent on a combination of both time and temperature).

The temperature compensation system may take into account long-term average temperature averages or trends. For example, a long-term average may be used to compensate for temperature fluctuations. A long-term average may take into account, for example, variations in body or skin temperature between the host and a baseline. In some examples, a long-term average method may be used in conjunction with one or more of the short-term (e.g., real-time) temperature compensation methods described below.

The average subcutaneous temperature for an individual may be determined and updated in a number of different ways. For example, the subcutaneous temperature may be determined as an average (e.g., mean or median) over the entire sensor session, or as an average over a rolling window (e.g., the last 12 or 24 hours). In some examples, the subcutaneous temperature may be updated at intervals, e.g., re-measured or updated every 6, 9, 12, 18, or 24 hours. In some examples, the subcutaneous temperature may be determined as a weighted average, with more recent values (e.g., the last 6, 12, or 24 hours) being weighted more heavily and older intervals being weighted more lightly.

In one example, the temperature sensor may be initially calibrated to an initial reference value (e.g., 35°C), which may represent the average temperature of the population. During a learning period, the temperature sensor may determine the actual temperature of the host. The learning period may be selected to be long enough (e.g., 6-12 hours) to filter out temperature excursions (e.g., so that the average is not determined during hot/cold events such as showers). The learned average may be used to compensate for host temperatures that differ from the population average. For example, if a population is assumed to have an operating temperature of 35.0°C, but temperatures detected from a particular host average a temperature of 35.5°C, a 1/2 degree variation may be used to compensate the analyte value. In some examples, an initial average may be determined (e.g., on the first day), and a provisional average may be updated for subsequent temperature measurements (e.g., using the average temperature on the second day, or the average temperature over two days). As explained above, other time windows may also be used.

If the system has a subcutaneous temperature sensor, a series of temperature measurements may be taken from the subcutaneous temperature sensor and used to determine a long-term average. In other examples, the subcutaneous temperature may be determined based on the sensed non-subcutaneous temperature (e.g., based on a linear or higher level relationship) using one of the various methods described below. After the subcutaneous temperature (Tsubcu,ind) for the individual is established, the temperature-corrected analyte sensitivity may be determined based on the deviation from the reference temperature (Tsubcu,reference), for example, using the formula provided above.

In some examples, the rate of change of the estimated glucose value or the rate of change of the signal from the glucose sensor may be used as an input to determine temperature compensation. For example, when the rate of change meets a condition (e.g., exceeds a specified value), temperature compensation may be paused or temperature compensation may be shifted to a different model. In some examples, when the rate of change of the estimated glucose value meets a condition, it may trigger an exception that can be addressed, for example, as described herein with respect to FIG. 34.

For some subcutaneous glucose sensors, the glucose concentration level determined from the subcutaneous sensor reflects a time delay relative to the blood glucose level due to physiological delays in changes in glucose levels in the interstitial fluid compared to changes in the blood (e.g., it may take several minutes for a change in blood glucose level to be reflected in the interstitial fluid measured by the subcutaneous glucose sensor). Delays may also be introduced by the periodicity of the sensor readings (e.g., if sensor readings are taken every 5 minutes, the estimated glucose level may be 4+ minutes out of date at some point in the cycle). When time lag errors are present in the system during times of fast glucose change, temperature compensation may be performed on the inaccurate (outdated) estimate of glucose. In some instances, temperature compensation on outdated glucose levels may deteriorate the estimate, and therefore it may be useful to pause or change temperature compensation during periods of high rates of change. For example, when the glucose concentration level is dropping rapidly (e.g., due to vigorous exercise), the estimate from the subcutaneous temperature sensor may "lag" behind the blood glucose concentration level (e.g., as determined from a blood glucose meter), and thus the subcutaneous sensor will show a higher estimated glucose value than the blood glucose level. If temperature compensation increases the estimated blood glucose concentration level of the subcutaneous sensor, it may exacerbate this discrepancy. This may be avoided by pausing temperature compensation or shifting to a different model. In some examples, when a high rate of change condition is met, temperature compensation may be applied only when temperature compensation increases the rate of change (e.g., to avoid exacerbating discrepancies caused by physiological delays).

In some examples, deviations of the analyte sensor output relative to the temperature sensor output may be used to evaluate the signal from the temperature sensor. These correlations or deviations may be used to establish confidence in the temperature signal, the analyte sensor signal, or both. During times when the glucose level meets a stability condition, the temperature and glucose concentration levels may be expected to show a correlation. The stability condition may be determined, for example, based on the rate of change of the glucose concentration level. In some examples, the stability condition may include multiple auxiliary conditions, such as short-term and long-term conditions. For example, a glucose level may be considered stable when the rate of change and/or the average rate of change over a specified period of time meets a condition (e.g., does not increase or decrease by more than 1 mg/dL per minute and/or does not increase or decrease by more than 15 mg/dL in 15 minutes). Glucose levels may be considered moderately stable (e.g., shown to be rising or falling at a moderate rate) if the rate of change and/or average rate of change or specified time period meets the conditions (e.g., glucose levels rise (or fall) 1-2 mg/dL per minute and/or rise (or fall) 15-30 mg/dL in 15 minutes).

As illustrated in Figures 15A-15C, the slopes of the temperature and glucose curves should correlate when the glucose level is stable (or in some instances, moderately stable). This is because the changes in the glucose curve reflect temperature-induced variations in the analyte sensor output. Figure 15A shows the glucose sensor output plotted against time. The gain in mg/dL is relatively high, showing the variation in slope during periods of relative glucose stability. Figure 15B shows the temperature sensor output plotted against time. Figure 15C shows the superposition of temperature on the glucose sensor output (i.e., the combination of Figures 15A and 15B). The analyte sensor output correlates with the temperature sensor output. The analyte sensor has an ascending value (positive slope) when the temperature sensor is ascending, the analyte sensor has a descending value (negative slope) when the temperature sensor output is descending, and the analyte sensor value is flat when the temperature sensor output is flat. The reliability of the temperature signal can be inferred from this correlation. In contrast, FIG. 15D shows an example where the temperature sensor output (dotted line) does not correlate well with the glucose sensor output during times of relatively stable glucose values, suggesting that the temperature sensor output may be unreliable.

In various examples, when confidence in the temperature sensor output is low, temperature compensation may be paused, reduced, or modified, or other information (e.g., motion detection as described below) may be used or sought to increase the accuracy of the temperature compensation. For example, if the confidence in the temperature signal is below a threshold, the analyte sensor may set an exception flag, which may be addressed as described herein with respect to FIG. 34.

FIG. 7 is a flow chart diagram of an example method 700 for determining a temperature-compensated glucose concentration level based on an evaluated (e.g., confirmed) temperature value. The method 700 may include, in operation 702, receiving a glucose sensor signal. For example, the glucose sensor signal may be received from a continuous glucose monitor (CGM).

The method 700 may include, at operation 704, receiving a temperature parameter signal. Receiving the temperature parameter signal may include, for example, receiving a signal indicative of a temperature, a temperature change, or a temperature offset.

Method 700 may include receiving a third sensor signal at operation 706. Receiving the third sensor signal may include, for example, receiving a heart rate signal, receiving a blood pressure signal, receiving an activity signal or an accelerometer signal (e.g., to detect motion), or receiving a location signal (e.g., to infer proximity to a hot or cold environment, such as a pool, beach, or air conditioning unit). In some examples, receiving the third sensor signal may include receiving temperature information from an ambient temperature sensor. In some examples, receiving the third sensor signal may include receiving information from a wearable device, such as a watch. In some examples, receiving the third sensor signal may include receiving temperature information from a physiological temperature sensor, which may be incorporated into, for example, a watch or other wearable device. In some examples, the third signal may include a heart rate signal, a respiration signal, a blood pressure signal, or an activity signal, and an exercise condition or state may be detected from an increase in the heart rate signal, the respiration signal, the blood pressure signal, or the activity signal.

The method 700 may include, at operation 708, evaluating the temperature parameter signal using the third sensor signal to generate an evaluated temperature parameter signal. In some examples, evaluating the temperature parameter signal may include determining a presence at a location having known temperature characteristics. For example, the low or high temperature signal may be corroborated by a location signal indicating a presence at a location having known ambient temperature characteristics (e.g., a hot or cold environment), such as a pool, a beach, an air conditioning unit, or an area with known weather characteristics, where the known weather characteristics may be determined, for example, by referencing a network resource (e.g., a website) or a stored lookup table. In some examples, the method may include determining a presence at a location having a submerged environment, such as a pool or a beach. In some examples, evaluating the temperature parameter signal may include determining that a change in the temperature parameter signal is consistent with an exercise session. For example, evaluating the temperature parameter signal may include determining that the temperature parameter signal is consistent with the occurrence of an elevated body temperature due to exercise.

The method 700 may include, at operation 710, determining a temperature-compensated glucose concentration level based on the evaluated temperature parameter signal and the glucose sensor signal. In some examples, determining the temperature-compensated glucose concentration level may include applying the temperature parameter signal to an exercise model. In some examples, the method may include using an exercise model (e.g., an outdoor or convection-cooled exercise model) when exercise is detected and a change in the temperature parameter signal indicates a decrease in temperature. For example, temperature compensation based on a non-subcutaneous temperature sensor (e.g., in the sensor electronics) may be suspended when the detected temperature decreases but exercise is detected (e.g., when HR or activity increases). This is because when a patient performs vigorous exercise (e.g., running) outdoors in a cold environment (e.g., when convection-cooled by a fan or exercising outdoors in a cold weather environment), the subcutaneous temperature may stabilize or even increase during the exercise session.

FIG. 8 is a schematic diagram of an example method 800 for temperature compensating a continuous glucose sensor that includes determining a pattern from temperature information. The method 800 may include, in operation 802, determining a pattern from the temperature data. In some examples, determining the pattern may include determining a pattern of temperature fluctuations, and the method may include compensating the glucose concentration level according to the pattern.

The method 800 may include, in operation 804, receiving a glucose signal from the continuous glucose sensor indicating a glucose concentration level.

The method 800 may include, at operation 806, determining a temperature-compensated glucose concentration level based at least in part on the glucose signal and the pattern. For example, the method may include receiving a temperature parameter, comparing the temperature parameter to the pattern, and determining a temperature-compensated glucose concentration level based at least in part on the comparison. In some examples, the pattern may include a temperature pattern that correlates with a physiological cycle, such as a circadian rhythm. In some examples, the method 800 may include determining whether the temperature parameter is highly reliable based on the comparison to the pattern, and temperature-compensating the glucose concentration level using the temperature parameter when the temperature parameter is determined to be highly reliable.

In some examples, the degree of compensation may be determined based at least in part on a comparison of the temperature parameter to the pattern. For example, the degree of compensation may be based on a defined range or confidence interval.

In some examples, the pattern may be determined by determining a condition or state, and determining the temperature-compensated glucose concentration level may be based at least in part on the determined condition or state. For example, the method 800 may further include receiving a temperature parameter, and determining the condition or state may include applying the temperature parameter to a state model. In some examples, determining the condition or state may include applying one or more of a glucose concentration level, carbohydrate sensitivity, time, activity, heart rate, respiration rate, posture, insulin delivery amount, meal time, or meal size to a state model. In some examples, determining the condition or state may include determining an exercise condition or state, and the method may include adjusting the temperature compensation-based model based on the exercise condition or state.

In some examples, a model may be selected or modified based on a detected condition or state. For example, a group of different linear models may be developed and a model may be selected from the group based on a detected condition or state of an analyte sensor, host, etc. In some examples, a condition or state may be determined using a state model.

In some examples, the condition or state may include a location or geographic characteristic. The location may include, for example, geographic location parameters (e.g., longitude, latitude, or altitude), or a city or place or point of interest (e.g., a beach or mountain). In various examples, location information, geographic information, or physiological sensor information (e.g., activity or heart rate as described below) may be collected from a host smart device, such as a cell phone, watch, or other wearable sensor.

In some examples, the condition or state may include a deviation of the temperature reading from the average. For example, a rolling average temperature value and a rolling standard deviation may be determined from the series of temperature values, and a model (e.g., a linear model) may be used depending on whether the temperature is +1σ, -1σ, +2σ, -2σ, +3σ, or -3σ away from the average. In some examples, the rolling average may be determined from a predetermined number of immediately preceding temperature values. In various examples, the current reading may be included or excluded from the rolling average. In some examples, the rolling average may be exponentially weighted.

In some examples, the condition or state may include patient demographics. For example, demographics may include gender (e.g., using different models for male vs. female hosts/patients), diagnosis (e.g., type 1 diabetes or type 2 diabetes or non-diabetic), age (e.g., age or youth, adolescent, adult, elderly), biological cycle (e.g., circadian rhythm or menstrual cycle), medical condition (e.g., pregnancy or health/illness or chronic disease).

In some examples, the condition or state may be determined from a wearable sensor or a physiological sensor, such as a heart rate sensor, an accelerometer, a blood pressure sensor, or a temperature sensor. The condition or state may be determined based on one or more sensor inputs. In some examples, the condition or state may include an activity condition or state, which may be determined, for example, from a heart rate or an accelerometer. In some examples, the condition or state may include an awake-sleep state, which may be determined from one or more physiological sensors (e.g., based on biorhythms) or from a posture sensor (e.g., a three-axis accelerometer). In some examples, the condition or state may include a compression state, which may be determined, for example, from a blood pressure sensor or a temperature sensor, or a combination thereof. For example, when a patient lies on a wearable glucose sensor, such as may occur during sleep, the sensor may generate an inaccurate glucose sensor reading (e.g., "low compression" indicating a lower glucose value). In some examples, each of these inputs or states may trigger a different temperature relationship (e.g., application of a particular temperature compensation model).

In some examples, temperature compensation, or its application (or suspension), may be based at least in part on a rate of change of temperature (e.g., the condition or state may be a rate of change of temperature). Because external (e.g., sensor electronics) detected temperature may change much faster than subcutaneous temperature, it may be difficult to accurately predict subcutaneous temperature during times of rapid temperature change. In some examples, temperature compensation may be suspended or reduced when the detected rate of change of temperature meets (e.g., satisfies) a condition (e.g., the rate of change exceeds a specified value, the rate of change falls outside a range, etc.). In another example, a first model (e.g., a linear model) may be used when a first condition is met (e.g., the temperature rate of change is below a specified value) and a second model (e.g., a linear delay model) may be used when a second condition is met (e.g., the rate of change is above a specified value).

In some examples, the temperature compensation, or its application (or suspension) may be based on the magnitude of the temperature gradient, also referred to herein as heat flux. The temperature gradient may be determined (e.g., approximated) from a determined subcutaneous or other in vivo temperature (e.g., a previous temperature determination) and a detected non-subcutaneous or ex vivo detected temperature, e.g., a temperature sensed at the sensor electronics. In one example, when a temperature gradient condition is met (e.g., a temperature gradient above a threshold), a model may be adjusted to reflect, for example, that the external temperature is changing faster than the subcutaneous temperature. In one example, a gain in a linear model (as described herein) may be reduced, which may have the effect of reducing the rate of change of the determined subcutaneous temperature to more accurately track the actual temperature change rate. In another example, temperature compensation may be suspended when a temperature gradient or heat flux condition is met. In some examples, the temperature compensation, or its application, may be based on the temperature gradient direction, e.g., temperature compensation when the external temperature is higher than the subcutaneous temperature may be different than temperature compensation when the external temperature is lower than the determined subcutaneous temperature.

In some examples, the condition or state may be exercise. For example, one or more wearable sensors (e.g., accelerometer, heart rate sensor, respiration sensor) may be used to determine whether the subject is engaged in some type of aerobic exercise (e.g., running, biking, or metabolic conditioning). In one example, it may be assumed that the subject's core body temperature (and subcutaneous temperature) is increasing, for example, from 37° C. to 38° C. During mobile exercise such as running or biking, it may also be assumed that the convection coefficient increases (e.g., by a factor of 10) due to the subject's movement. An appropriate exercise model that accounts for these parameter changes may be applied. For example, the "gain" (slope) of the linear model may be increased and the offset (constant) may be changed to reflect the effect of exercise (e.g., a basic linear model: Tsubcu=0.395 ^* Texternal+22.346 may be shifted to an aerobic linear model such as Tsubcu=0.416 ^* Texternal+22.178).

In some examples, the amount of temperature compensation applied may be limited when cold temperatures and motion are detected. For example, during motion, the transmitter temperature may be lower than when the subject is at rest, for example, because the subject is outdoors, the sensor electronics are exposed to increased convection due to motion, or the subject's skin becomes cold due to sweating. However, the subcutaneous temperature may rise due to increased heat generation, so standard temperature compensation models (which do not consider the motion/cold combination) may result in inaccuracies. This "cold motion" condition or state may be detected, for example, through a combination of temperature sensor input and accelerometer, heart rate, respiration rate, or position input. When motion by the subject is detected, the temperature compensation may be modified to a lower temperature, for example, temperature compensation may be paused, limited, or tapered, or an alternative compensation model may be applied. In one example, during motion, any temperature below a threshold may be treated as a threshold for purposes of temperature compensation (e.g., a sensor temperature < 29°C is replaced with 29°C for purposes of temperature compensation), or temperature compensation may be limited by an algorithm. In another example, taper compensation may be achieved by decreasing the temperature sensitivity coefficient (Z) so that for a given detected sensor temperature, a smaller change is made to the subcutaneous temperature (e.g., Mt,comp=Mt,pro ^* (Z) ^* (Tsubcu-Tsubcu,reference), or Mt,comp=Mt,pro ^* (Z) ^* (Tsubcu-Tsubcu,reference)+Mt,pro). For example, if a typical temperature sensitivity coefficient is 3.3%, during exercise the temperature sensitivity coefficient may be changed to 1.5% based on the detected condition or state.

In some examples, the compensation model may be selected or determined based at least in part on the individual's average subcutaneous temperature. In one example, the average subcutaneous temperature may be established for the first few hours of the session (e.g., during or after the warm-up period) or may be established for the first day of the session. The long-term averaging method described above may be used to determine the compensation. In some examples, the average subcutaneous temperature may be updated periodically or recurringly, for example, every 6 hours, every 12 hours, or every 24 hours.

In some examples, a determination may be made (e.g., using an algorithm, model, or lookup table) as to whether temperature compensation may increase the accuracy of the estimated glucose value. For some patients or conditions, temperature compensation may actually decrease accuracy. Identifying these patients or discriminators and pausing or withholding temperature compensation may improve sensor performance or decrease MARD. Discriminators may include, for example, host surface or body temperature, BMI, sex, age, or any combination of the other conditions or states identified above.

FIG. 9 is a flow chart diagram of an example method 900 for temperature compensating a continuous glucose monitoring system based at least in part on a detected condition or state. The method 900 may include, at operation 902, receiving a glucose signal indicative of a glucose concentration level.

The method 900 may include, at operation 904, receiving a temperature signal indicative of a temperature parameter. The method 900 may include, at operation 906, detecting a condition or state. In some examples, the condition or state may include a high rate of change of the glucose signal, and temperature compensation may be reduced or suspended during periods when the glucose signal is undergoing a high rate of change. In some examples, the condition or state may include a body mass index (BMI) value. For example, it may be assumed that a host having a high BMI naturally warms up or changes temperature more slowly than a host having a low BMI. In some examples, the condition or state may include a detected fever, and temperature compensation may be reduced, suspended, limited, or tapered in response to the detection of the fever. In some examples, the detected condition or state may include the presence of radiant heat on the continuous glucose monitoring system. In some examples, the condition or state may include detected movement. The method may include, for example, decreasing, tapering, limiting, or pausing temperature compensation when a condition or state (e.g., movement) is detected.

In some examples, the glucose signal may be received from a continuous glucose sensor, and the condition or state may include pressure on the continuous glucose sensor. For example, pressure on the sensor may be detected based at least in part on a rapid drop in the glucose signal. In some examples, the condition or state may include sleep. In some examples, the condition or state may include pressure during sleep. Sleep may be detected using, for example, one or more of temperature, posture, activity, and heart rate, and the method may include applying a designated glucose alert trigger based on the detected sleep.

The method 900 may include, at operation 908, determining a temperature-compensated glucose concentration level based at least in part on the glucose signal, the temperature signal, and the detected condition or state.

In some examples, the condition or state may include a sudden change in the temperature signal. Temperature compensation may be reduced or suspended, for example, in response to detection of a sudden change in temperature. A sudden change in temperature is unlikely to occur at the analyte sensor site in a subcutaneous location, where temperature changes tend to occur more gradually as heat is conducted through the skin to or away from the sensor site; therefore, when a sudden temperature change occurs at the external sensor, it may be appropriate to perform a responsive action, such as, for example, suspending temperature compensation for a period of time, or "stepping" temperature compensation over a period of time to reflect the gradual temperature change at the sensor site.

After a sudden change in temperature or other rapid change or signal discontinuity is detected, one or more of a variety of techniques may be used to determine the temperature-compensated glucose level. In some examples, the temperature-compensated glucose concentration level may be determined using a previous temperature signal value in place of the temperature signal value associated with the sudden change in temperature. In some examples, the temperature-compensated glucose concentration level may be determined using an extrapolated temperature signal value based on the previous temperature signal value and using the extrapolated temperature signal value in place of the temperature signal value associated with the sudden change in temperature. In some examples, a delay model may be invoked in response to detecting the sudden change in temperature. For example, the delay model may specify a delay period for use in determining the temperature-compensated glucose level.

One or more of a variety of techniques may be used to determine the temperature-compensated glucose concentration level based on the glucose signal, the temperature signal, and the detected condition or state. For example, a linear model may be used to determine the temperature-compensated glucose concentration level. In another example, a time series model may be used to determine the temperature-compensated glucose concentration level. In some examples, a partial differential equation may be used to determine the temperature-compensated glucose concentration level. In some examples, a probabilistic model may be used to determine the temperature-compensated glucose concentration level. For example, a state model may be used to determine the temperature-compensated glucose concentration level.

In some examples, the method may include determining a long-term average using the temperature signal, and a temperature-compensated glucose concentration level may be determined using the long-term average.

In some examples, the method may further include receiving a blood glucose calibration value and updating the temperature compensation gain and offset when the blood glucose calibration value is received.

The method may further include delivering insulin therapy (e.g., via a pump or smart pen). The insulin therapy may be determined based at least in part on the temperature-compensated glucose level.

In some examples, the temperature compensation may be based at least in part on a body mass index (BMI). In one example, height and weight may be received from the subject, for example, via a smartphone application interface. The temperature compensation parameters may be determined or adjusted based at least in part on the BMI. In some examples, the temperature compensation may be based on a preloaded model, which may be associated with a specified BMI window. For example, a standard temperature compensation model may assume a specific distance from the subcutaneous layer (where the working electrode is designed to be located during use) and tissue at the core body temperature. In a person with a high BMI, a thicker layer of adipose tissue (body fat) may extend the distance from the subcutaneous layer to tissue at the core body temperature, thereby resulting in a lower subcutaneous or skin surface temperature. In some examples, a group of models may be available, and the model (e.g., a PDE model with varying distance to the core body temperature) and the model may be selected from the group based at least in part on the person's BMI. In some instances, because BMI does not perfectly predict adipose tissue thickness, especially at the location of the analyte sensor (e.g., CGM), additional information may be used in addition to BMI to select the model.

In some examples, the temperature compensation model is based at least in part on the subject's core body temperature. For example, body temperature tends to correlate with BMI, so the average body temperature may be estimated based on BMI.

Other physiological factors or influences, such as local glucose concentration fluctuations (as opposed to systemic glucose levels), compartment bias (differences in glucose concentrations between interstitial fluid and blood), and non-enzymatic sensor biases, may also be considered in determining the compensated analyte value.

In some examples, a sensor signal from an optical sensor having a light source and an optical sensor can be input for the temperature compensation method. For example, an optical sensor can be used to detect blood flow or perfusion at the skin of a subject. An optical sensor having a light source and a photodetector near the skin of a subject can detect blood flow velocity and red blood cell count in the area directly under the sensor. Blood flow near the skin varies with temperature, activity, and stress level. In some examples, the amount of exertion (e.g., exercise) can be determined by using blood perfusion information obtained from the optical sensor. For example, when running uphill or downhill, the number of steps is approximately the same, but the uphill requires more exercise and the blood perfusion is higher. During the downhill section, the blood perfusion is lower. Specific motion detection can be used for more sophisticated temperature compensation algorithms used during exercise. In some examples, an optical sensor may detect less obvious exercise than an accelerometer (e.g., weight training) because the exercise involves less or slower movement. In some examples, optical sensors may be used in conjunction with accelerometers to detect motion conditions or states and amounts of motion during exercise.

In some examples, location information (e.g., global positioning sensor data or network connectivity) may be used as an input to determine temperature compensation or to determine the confidence of a temperature measurement. For example, location may be used to establish the confidence of a temperature measurement by comparing the temperature measurement to the temperature characteristics of the location. For example, an activity associated with a location (e.g., swimming, sunbathing, running, skiing) may establish confidence in a low temperature measurement, a high temperature measurement, or a rapidly changing temperature measurement. In another example, weather characteristics at a location (e.g., ambient temperature) may establish confidence in a temperature measurement. In another example, a temperature measurement may be confirmed using location information that correlates with a circadian rhythm (e.g., typically sleeps at a home location), or deviations from a circadian rhythm may be confirmed by deviations from a pattern in the location information (e.g., confidence in low temperatures at night may be established if the subject is away from home, e.g., outdoors at night, camping, in a location that may have different temperature characteristics).

In some examples, detection of a fever (e.g., using a sensor) or reporting of a fever (e.g., through an app on a smart device) may be used as an input to determine temperature compensation. For example, temperature compensation may be paused during a fever since normal patterns may not apply. In another example, models may be modified or different models may be applied to compensate for temperature changes caused by a fever. In some examples, a fever may be corroborated using other information. For example, correlations of the rate of change of the sensor output illustrated in Figures 15A-15C may be used to corroborate a detected fever. In another example, the patient may be queried, for example, by the smart device with queries regarding fever ("Do you have a fever?") or other events that may cause a temperature change ("Have you taken a bath recently").

Figure 19 is a schematic diagram of an exemplary model that may be used to determine an output from two or more inputs. For example, the model may learn patterns or relationships from previous data and apply the learned patterns or relationships in determining the output. This may include, for example, learning from previous data from a particular host, or population, or one or more clinical trials.

In various examples, the inputs may be received or sensed simultaneously or at different times. In some examples, two inputs (e.g., temperature and analyte sensor output) may be applied to the model. The model may also receive additional inputs, such as time (e.g., received from a clock circuit) or sensitivity (e.g., factory calibrated sensitivity). In one example, the model may include sub-models 1902, 1904, 1906. The sub-models may consider temperature-dependent factors such as local glucose levels, compartment bias values, non-enzymatic sensor bias levels, and sensor sensitivity. In one example, each model may define a different relationship (e.g., linear, non-linear) between the inputs and the temperature-dependent factors. For example, model 1902 may be based on a first non-linear relationship, model 1904 may be based on a second non-linear relationship, and the output model may be based on a linear relationship. In various examples, the processor may retrieve model information or input data from a lookup table in memory, or may store and retrieve past values or states to memory, or may retrieve functions or other aspects of the model from memory. The retrieved information may be combined with recent or real-time information and applied to a model to generate an output that may be a compensated glucose concentration level, or the output may be used to determine a compensated glucose concentration level.

FIG. 20A is a flow chart diagram of an example method 2000 of determining a compensated glucose concentration value using a model. The method 2000 may include receiving a temperature sensor signal in operation 2002. For example, the temperature sensor signal may be received from a subcutaneous temperature sensor proximate to the analyte sensor, or the temperature sensor signal may be received from a non-subcutaneous sensor (e.g., on external sensor electronics, e.g., a CGM transmitter). In operation 2004, the method 2000 may include receiving an analyte sensor signal, such as a signal from a glucose sensor. In operation 2006, the temperature sensor signal and the glucose sensor signal may be applied to a model. For example, the temperature sensor signal and the glucose sensor signal may be applied to a state model (e.g., a hidden Markov model) or a neural network. In some examples, multiple temperature sensor signals may be applied to the model. The signals may be processed or analyzed to determine a pattern (e.g., one or more linear or nonlinear trends). The defined or learned relationship between the temperature and glucose sensor values and the compensated glucose concentration value may be used to return a compensated glucose concentration value using the model. In operation 2008, the compensated glucose concentration value may optionally be displayed on the user device. In operation 2010, a therapy may be delivered based at least in part on the compensated glucose concentration value. For example, the amount of insulin delivered via the pump may be controlled based at least in part on the compensated glucose concentration value. In some examples, the processor may determine an insulin dose, delivery time, or delivery rate (or any combination thereof) based at least in part on the glucose concentration value. In some examples, the pump may automatically deliver insulin, or the pump may suggest insulin time, rate, and dosage to the user. In other examples, a smart pen may receive the compensated glucose concentration value and determine a dose or delivery time, which may be displayed to the user, automatically loaded for delivery, or both.

In the example of FIG. 20A, the model is trained to provide as its output a compensated glucose concentration. In other examples, the model is trained to generate an output that includes one or more compensated characteristics of the glucose sensor, as described herein. For example, as described herein, temperature compensation may be applied to the sensor characteristics to generate one or more compensated sensor characteristics. The one or more compensated sensor characteristics may then be applied to the raw sensor data to generate a compensated glucose concentration. Exemplary sensor characteristics that may be compensated using the trained model include, for example, sensitivity, sensor baseline, etc.

20B is a flow chart diagram of another example method 2001 of determining a compensated glucose concentration value using a model. Method 2001 may include receiving a temperature sensor signal at operation 2012. For example, the temperature sensor signal may be received from a subcutaneous temperature sensor proximate to the glucose sensor, or the temperature sensor signal may be received from a non-subcutaneous sensor (e.g., on external sensor electronics, e.g., a CGM transmitter). At operation 2014, method 2001 may include receiving a glucose sensor signal, such as a signal from a glucose sensor. The glucose sensor signal received at operation 2014, in some examples, includes a raw sensor signal related to the current at the working electrode, such as one or more counts related to the current at the working electrode of the glucose sensor. In some examples, this includes an estimated analyte value, such as a glucose concentration, derived from the raw sensor signal. In some examples, the glucose sensor signal includes the raw sensor signal and an analyte value.

In operation 2016, the temperature sensor signal and the glucose sensor signal may be applied to a model. For example, the temperature sensor signal and the glucose sensor signal may be applied to a state model (e.g., a hidden Markov model), a neural network model, or other suitable model. In some examples, multiple temperature sensor signals may be applied to the model. The signals may be processed or analyzed to determine patterns (e.g., one or more linear or non-linear trends). Defined or learned relationships between the temperature and glucose sensor values and one or more glucose sensor characteristics, such as sensitivity, baseline, etc., may be used to return values of one or more compensated glucose sensor characteristics.

In operation 2018, the compensated glucose sensor characteristics are used to generate a compensated glucose concentration. In operation 2020, the compensated glucose concentration value may optionally be displayed on the user device. In operation 2022, a therapy may be delivered based at least in part on the compensated glucose concentration value. For example, the amount of insulin delivered via the pump may be controlled based at least in part on the compensated glucose concentration value. In some examples, the processor may determine an insulin dose, delivery time, or delivery rate (or any combination thereof) based at least in part on the glucose concentration value. In some examples, the pump may automatically deliver insulin, or the pump may suggest insulin time, rate, and dosage to the user. In other examples, the smart pen may receive the compensated glucose concentration value and determine a dose or delivery time, which may be displayed to the user, automatically loaded for delivery, or both.

In some examples, temperature compensation or estimated subcutaneous temperature may be based at least in part on the electrical conductance (or electrical resistance, the inverse of conductance) of the analyte sensor or a portion thereof. For example, the measured conductance of the analyte sensor 10 or conductive portion 286 of the analyte sensor shown in FIG. 2C may be used for temperature compensation or estimation of the subcutaneous temperature.

Empirical measurements (discussed below and shown in FIG. 21) indicate that the conductance of an analyte sensor can be strongly dependent on temperature. In some examples, this relationship between conductance and temperature can be used to estimate the subcutaneous temperature, which can be used in a temperature compensation model or other method. In other examples, the relationship between conductance and temperature can be applied directly (e.g., without using an estimated temperature) to compensate for subcutaneous temperature variations.

Figure 21 is a plot of sensor conductance 2103 and transmitter temperature 2105 versus time. A strong correlation is seen between temperature and conductance. As transmitter temperature increases, sensor conductance increases (approximately 6% per degree Celsius) and vice versa. The data shown is for transmitter temperature, but the same correlation exists between subcutaneous temperature and conductance.

The correlation between temperature and sensor conductance may be used to determine an estimate of the temperature at the working electrode temperature (e.g., to determine the subcutaneous temperature at the analyte sensor). In various examples, the system or method may use a non-subcutaneous temperature (e.g., the transmitter temperature), or the system or method may compensate without using a non-subcutaneous temperature (e.g., the system may use an assumed reference temperature or a factory calibrated temperature, as described above).

An initial estimate for the working electrode temperature may be made using one (or more) of a variety of models (e.g., linear model, delay model, partial differential equation model, time series model). The initial estimate may also be based on a predetermined reference value, or other methods described herein. This initial estimate may then be used to determine an adjusted temperature using one or more sensor conductance measurements. For example, as the conductance changes, a corresponding temperature change may be calculated, and this temperature change may be applied (e.g., added to or subtracted from) the initial temperature estimate or reference temperature to determine the temperature at the time of the sensor conductance measurement.

In various examples, the conductance-based temperature compensation technique may be combined with any of the examples described herein to determine an estimated subcutaneous temperature or an estimated effect of subcutaneous temperature on a signal from an analyte sensor. For example, an estimated subcutaneous temperature (e.g., the temperature at the working electrode of the analyte sensor) may be determined from a non-subcutaneous temperature (e.g., the transmitter temperature) measured at a first time, and the conductance of the analyte sensor or a portion thereof may be measured simultaneously with the non-subcutaneous temperature measurement. A second subcutaneous temperature may then be estimated based on the difference between the subsequent conductance value (single point or average) and the conductance value (single point or average) from the first time.

The conductance values 2013 plotted in FIG. 21 show an upward drift over time. This drift component can be related to sensor sensitivity drift, as described in U.S. Patent Application Publication No. 2015/0351672, which is incorporated by reference.

In some examples, the system may implement one or more techniques to account for drift and avoid or reduce the effects of conductance drift on the subcutaneous temperature estimate or compensated data. Such techniques to address drift may be, for example, resetting the temperature estimate (e.g., recalculating the estimated temperature and the conductance baseline used in compensating future values), average-based compensation (e.g., compensation for a moving baseline conductance based on a long-term average, a weighted average, or a rolling window).

In various examples, the subcutaneous temperature estimate or conductance baseline may be periodically refreshed. For example, a new subcutaneous temperature estimate (e.g., working electrode temperature) may be iteratively (e.g., periodically) refreshed (e.g., reset) by recalculating the estimate (e.g., using the techniques discussed above). Future analyte values may be compensated for a conductance value (or average) that is time-correlated (e.g., contemporaneous) with the new subcutaneous temperature estimate. This refreshing (resetting) of the conductance-based temperature estimate may eliminate or reduce the effects of the drift component, resulting in a more accurate temperature estimate.

In some examples, a reset, refresh, or error status may be triggered based on a condition being met. For example, meeting a condition may trigger an exception and response as described herein with respect to FIG. 34. The condition may be based on a comparative conductance-compensated temperature estimate, including, for example, a subcutaneous temperature estimate determined in a different manner (e.g., not based on conductance), such as a transmitter temperature and a newly calculated subcutaneous temperature estimate based on a linear model, a delay model, or other models discussed herein. The condition may be met, for example, when the two values differ by more than a set threshold. In some examples, when the comparison meets an error condition, an error status may be changed (e.g., an error condition or state may be declared. In some examples, the conductance baseline may be reset (e.g., the baseline may be updated to a new value or average) or a new temperature estimate may be correlated with a particular conductance value. In some examples, a stepwise approach may be applied such that when the difference exceeds a reset threshold, a reset procedure may be applied, and when the difference exceeds an error threshold that is greater than the reset threshold, an error condition may be applied (in which case a reset may or may not still occur). Resetting the conductance-based temperature estimate in this manner may eliminate or reduce the drift component visible in the conductance signal of FIG. 21 (e.g., the conductance value drifts up over time).

In some examples, a digital high-pass filter may be applied to block low-frequency drift components from the conductance signal and pass only temperature-related changes. The filter characteristics, such as the cutoff frequency, may be based on measured temperature data, preferably subcutaneous temperature measurement data (e.g., by frequency analysis such as Fourier decomposition).

Although the above discussion has focused on conductance and resistance, it should be understood that temperature compensation or temperature estimation may alternatively be based on other electrical conductance properties (e.g., impedance or admittance) depending on the configuration of the analyte sensor system and the type of signal applied.

22 is a flow chart diagram of an exemplary method 2200 of temperature compensation using conductance or impedance. In operation 2202, a first value indicative of the conductance of the sensor component at a first time is determined. In operation 2204, a second value indicative of the conductance of the sensor component thereafter is determined. In operation 2206, a signal representative of an analyte concentration in the host is received. In operation 2208, a compensated analyte value is determined based at least in part on a comparison of the second value to the first value. In some examples, determining the first value may include determining an average conductance over a period of time that is proximate to or includes the first time. In some examples, the method may further include determining a first estimated subcutaneous temperature that is time-correlated with the first value and determining a second estimated subcutaneous temperature that is time-correlated with the second value, the second estimated subcutaneous temperature being determined based at least in part on a comparison of the second value to the first value.

In some examples, the method may include determining a third estimated subcutaneous temperature time-correlated with the second value, determining whether a condition is met based on a comparison of the third estimated subcutaneous temperature to the second estimated subcutaneous temperature, and setting an exception in response to meeting the condition. In some examples, setting the exception includes setting a flag or other indicator that may be treated as an exception in the manner described herein with respect to FIG. 34. For example, the exception may cause a response action, such as, for example, triggering a reset of the analyte sensor system. Triggering a reset of the analyte sensor system may include determining a subsequent estimated subcutaneous temperature based on the third estimated temperature and the second value, or based on a third value indicative of a subsequent conductance and a fourth estimated subcutaneous temperature time-correlated with the third value.

In some examples, method 2200 may include compensating for drift in the conductance values, for example, by applying the methods described above or by applying a filter.

FIG. 23 is a flow chart diagram of an example method 2300 for determining an estimated subcutaneous temperature using conductance or impedance. In operation 2302, a first value indicative of the conductance of the sensor component at a first time may be determined, for example, by a measured conductance or impedance of the sensor component. In operation 2304, a second value indicative of a subsequent conductance of the sensor component may be determined, for example, by taking a second measurement to determine the conductance or impedance. In operation 2306, an estimated subcutaneous temperature may be determined based at least in part on a comparison of the second value to the first value. As described above, an initial estimated temperature may be determined using a non-subcutaneous temperature measurement, and a subsequent estimated subcutaneous temperature may be determined based on a change in the value indicative of the conductance. If the variation exceeds a threshold or otherwise the comparison meets an error or reset condition, an error condition may be set or a reset may be triggered. In some examples, an error condition or reset condition, when detected, may be treated as an exception in the manner described herein with respect to FIG. It should be understood that any of the estimated temperatures described herein may be used as an input for any of the temperature compensation models described herein.

In some examples, the temperature sensor may be calibrated during a manufacturing step where the process temperature is known or controlled. For example, some sensor electronics packages use adhesives or structural agents such as epoxies that can be cured at a known or controlled temperature. The temperature sensor may be calibrated during the curing step. In another example, the temperature sensor may be calibrated when the analyte sensor is calibrated. In another example, the temperature sensor may be calibrated during an initial period of wear. For example, the temperature sensor output during the initial period (e.g., the first hour or two after start-up of the analyte sensor) may be calibrated to a predetermined average (e.g., 37°C).

FIG. 10 is a schematic diagram of a method 1000 for temperature compensating a continuous glucose sensor system using a reference temperature value. The method may include, in operation 1002, determining a first value from a first signal indicative of a temperature parameter of a component of the continuous glucose sensor system. The method may include, in operation 1004, receiving a glucose sensor signal indicative of a glucose concentration level. The method may include, in operation 1006, comparing the first value to a reference value.

The method may include, in operation 1008, determining a temperature compensated glucose level based on the glucose sensor signal and a comparison of the first signal to a reference value.

In some examples, the method may further include determining a reference value. For example, the reference value may be determined from the first signal. For example, the continuous glucose sensor system may include a glucose sensor insertable into a host, and the reference value may be determined during a particular period of time after insertion of the glucose sensor into the host or after activation of the glucose sensor. In other examples, the reference value may be determined during a manufacturing process.

In some examples, the reference value may be during a first period of time, and the first value may be determined during a second period of time after the first period of time (e.g., the reference value may be established after insertion of a sensor, and subsequent sensor readings may be compensated relative to the reference value). In some examples, the reference value may be a long-term average, and the first value may be a short-term average. In some examples, the reference value may be updated based on subsequently received temperature values. For example, the reference value may be updated based on one or more temperature signal values obtained in a third period of time after the second period of time.

In some examples, the reference value may be determined based on an average of multiple sample values obtained from the first signal.

11 is a flow chart diagram of an exemplary continuous glucose sensor temperature compensation method 1100. The method may include receiving a calibration value for a temperature signal in operation 1102. In some examples, the calibration value may be obtained during a manufacturing step having a known temperature. In some examples, the calibration value for the temperature signal may be obtained during a specified period of time after insertion of the continuous glucose sensor into the host. For example, the calibration value may be determined after a warm-up period, which may be, for example, a two-hour period after insertion or activation of the sensor. For example, the calibration value may be determined during a subsequent period of time after the warm-up period (e.g., 2-4 hours after insertion). The method may include receiving a temperature signal from the temperature sensor indicative of a temperature parameter in operation 1104. The method may include receiving a glucose signal from the continuous glucose sensor indicative of a glucose concentration level in operation 1106. The method may include determining a temperature-compensated glucose concentration level based at least in part on the glucose signal, the temperature signal, and the calibration value in operation 1108.

In some examples, relative temperature variation may be used for temperature compensation. For example, uncalibrated temperature sensors or temperatures with low absolute accuracy may be used for temperature compensation by basing temperature compensation on deviations from a reference, as opposed to knowledge of the absolute temperature. This may include, for example, using an individualized dynamic reference temperature (e.g., a reference temperature determined for a particular sensor or session that may be periodically refreshed or recalculated) and applying compensation using deviations from that reference temperature.

In some examples, the temperature difference may be determined from a reference condition or state based on the variation of the first value from a reference value without calibrating the temperature to the reference value. This may allow, for example, to compensate for the temperature difference from the reference value even when the absolute temperature is not determined. This may be useful to ensure accurate absolute temperatures when the temperature sensor is not factory calibrated, or may be useful when using a sensor that has good relative accuracy or precision, but less reliable absolute accuracy or precision. In some examples, the temperature-compensated glucose level may be determined based at least in part on a temperature-dependent sensitivity value that varies based on the deviation of the first value from the reference value.

In some examples, temperature compensation may be implemented using a temperature sensor with low absolute accuracy. For example, even if the sensor is not accurate in an absolute sense (e.g., ±3°C or 5°C variation in absolute temperature), the sensor may be sufficiently accurate in a relative sense (e.g., the sensor accurately detects that it is 1°C warmer than a previous (reference) point in time). The use of these types of sensors may be advantageous because the sensor may be built into the sensor electronics for other reasons (e.g., to detect overheating) and the calibration steps required may be simpler or less expensive.

In one example, the reference temperature may be obtained when a blood glucose value is received (e.g., a blood glucose meter using a finger prick). For example, when the blood glucose value is received, a glucose sensitivity may be determined (e.g., calculated) based on a signal from an analyte sensor (glucose sensor) and a signal from the temperature may be obtained (e.g., declared) as the reference temperature. The signal from the temperature sensor may then be used to determine a temperature difference from the reference temperature, and temperature compensation may be based on that difference. For example, the temperature may then be determined to be 1.5° C. higher than the reference temperature, and temperature compensation may be applied based on the 1.5° C. difference. In some examples, temperature compensation may be based on a raw or processed signal from the temperature sensor as opposed to a calculated temperature difference.

In various examples, the baseline temperature may be determined during a particular time period, for example, the first 2 hours or the first 24 hours after a sensor session is initiated. In one example, the baseline temperature may be the average (e.g., mean or median) temperature during the specified time period. In some examples, the baseline temperature may be used for the remainder of the session. In other examples, the baseline temperature may be updated repeatedly or periodically. For example, the baseline may be updated every 24 hours, and the baseline temperature may be used for the following 24 hours. In some examples, for temperature compensation purposes, the baseline temperature may be assumed to be a particular value (e.g., 35C may be assumed as the average subcutaneous temperature for the general population of subjects). In some examples, the baseline value may be the temperature sensor value at the time of calibration (during manufacture or after insertion).

Real-time temperature compensation may be determined using any of the compensation methods described herein (linear compensation, linear compensation with delay, polynomial compensation, etc.) using the real-time (or recent) temperature signal and a reference temperature value. In some examples, temperature compensation using relative temperature can obtain 75% (or more) of the MARD improvement achieved using a calibrated temperature sensor.

Indicators of movement or condition may be detected and used to determine temperature compensation. Movement may be detected, for example, based on temperature data, accelerometer data (e.g., to detect walking or running), location data (e.g., based on presence at a location associated with movement or based on movement at a location associated with walking, running, or biking), physiological data (e.g., respiration, heart rate, or skin surface condition).

In some examples, the method may include detecting an increase in the first temperature signal and a decrease in the second temperature signal, and adjusting a temperature compensation model based on the detected increase and decrease. In some examples, an exercise session (e.g., outdoor exercise or convection-cooled exercise) may be detected based at least in part on the detected increase in the first signal and the decrease in the second signal. For example, a decrease in the second signal may indicate the start of an exercise session in a cool environment (e.g., outdoors on a cold day, or an exercise session in an actively cooled environment (e.g., near a fan)). A decrease in the temperature signal from an external second sensor (e.g., in the sensor electronics) may indicate a decrease in temperature in response to an outdoor ambient temperature being lower than an indoor ambient temperature, or a decrease in temperature in response to convection cooling (e.g., from running or biking, or a fan adjacent to a treadmill or other training space, for example). For example, an increase in temperature (or a steady state temperature) in the first temperature signal, which may be received from an external sensor positioned closer to the body than the second sensor, or may be received from a sensor that is subcutaneous (e.g., on or integrated into the glucose sensor), may indicate no decrease in temperature despite a change in ambient temperature due to body warming from exercise or heat generation from exercise.

12 is a flow chart diagram of an example method 1200 of temperature compensation using two temperature sensors. Method 1200 may be implemented, for example, in the system shown in FIG. 2C. Method 1200 may include, in operation 1202, receiving a glucose signal from a glucose sensor, the glucose signal representing a glucose concentration level of the host.

The method 1200 may include, at operation 1204, receiving a first temperature signal indicative of a first temperature parameter proximate the host or the glucose sensor. The method 1200 may include, at operation 1206, receiving a second temperature signal indicative of a second temperature parameter. In some examples, the first temperature signal may be received from a first temperature sensor coupled to the glucose sensor and the second temperature signal may be received from a second temperature sensor coupled to the glucose sensor.

Method 1200 may include, at operation 1208, determining a compensated glucose concentration level based at least in part on the glucose signal, the first temperature signal, and the second temperature signal. In some examples, the compensated glucose concentration level may be determined based at least in part on a temperature gradient between the first temperature sensor and the second temperature sensor, or based at least in part on a heat flux between the first temperature sensor and the second temperature sensor. In some examples, method 1200 may include detecting an exercise session based on two temperature signals (e.g., based on a difference in the detected temperatures) and compensating accordingly (e.g., applying an exercise model), as described above.

In some examples, the method 1200 may further include determining that the temperature change is due to radiant heat or ambient heat based at least in part on the second temperature signal, and adjusting the temperature compensation model based on the determination. For example, if the second temperature signal is from a sensor near an exterior surface of the wearable sensor and the second temperature signal is significantly higher than the first temperature signal, it may be inferred that the sensor is exposed to radiant heat. In some examples, the rate of change may also be considered. For example, a rapid rate of change may indicate immersion in hot water, and a slower rate of change may indicate exposure to radiant heat. In some examples, the state model may include one or more of a radiant heat state, a submersion state, an exercise state, an ambient air temperature state, or an ambient water temperature state, and the state model may be used for temperature compensation of the estimated glucose value.

The temperature sensor may be used for a variety of other purposes. In some examples, BMI may be estimated from temperature. For example, lower temperatures tend to correlate with higher BMI. The estimated BMI value may be shared with other applications. For example, a decision support system may use the BMI as an input for a model or algorithm to determine guidance for a subject (e.g., a glucose correction dose, exercise recommendations, or eating a certain amount or type of carbohydrates or food).

In some examples, an alarm or alert may be triggered when a temperature sensor indicates a temperature that meets a condition. For example, an alarm or alert may be triggered when a temperature sensor indicates a temperature that meets a statistical condition (e.g., the temperature is more than one standard deviation away from a mean or baseline value, or is more than a specified number of standard deviations away from a mean or baseline value). For example, a potentially dangerous or hazardous condition or state of a patient (e.g., severe fever, heat stroke, hypothermia, etc.) may be detected using a subcutaneous temperature sensor or using a temperature sensor in the sensor electronics, and the condition or state may be communicated via an alarm or alert (e.g., via the subject's smart device or through a wireless network or the Internet to a caregiver's smart device). In other examples, a potentially overheated or excessively cold sensor or sensor electronics may be detected. In some examples, a potentially faulty temperature sensor may be identified based on a temperature sensor signal that meets a condition (e.g., when the temperature sensor indicates a temperature within an unlikely range).

In various examples, temperature compensation as described herein may be utilized in conjunction with analyte sensors for measuring analytes other than glucose. Temperature compensation techniques may be used with analyte sensors for measuring any analyte, including the exemplary analytes described herein.

Also, in some examples, temperature measured by a subcutaneous temperature sensor or using a temperature sensor in the sensor electronics as described herein may be used to determine a recommended insulin dosage. For example, the host's body may utilize insulin differently depending on temperature. Temperature-related adjustments may be made to the host's insulin dose based on the measured temperature.

Sensor disconnection or reuse of a disposable sensor ("restart") may be detected based at least in part on a temperature change or lack of temperature change. Some analyte-based sensor systems may be configured with a disposable (replaceable) sensor component and a reusable sensor electronics package, e.g., a CGM transmitter, that may be mechanically and electrically coupled to the disposable sensor component. The disposable sensor component may be designed to extend into the subcutaneous layer of the host and function for several days (e.g., 7, 10, or 14 days), after which the disposable sensor component is removed and replaced with a new disposable sensor component. As described in detail in the discussion of FIG. 1, the reusable transmitter may be wirelessly coupled to a control device (e.g., a smart device), which may include a user interface for entering commands that may be sent to the transmitter. The user interface on the control device may allow for stopping a sensor session and starting a new sensor session.

A sensor session can be programmed for a defined period of time (e.g., seven days), after which the session expires (unless manually stopped via the user interface). After a sensor session expires or is stopped, a new session can be started via the user interface.

In some cases, a subject (e.g., a patient) may begin a new sensor session without replacing the disposable sensor components, i.e., the subject may "restart" the session with the same disposable components that were used before the session was stopped. Detecting such restart events may be useful for a variety of reasons.

A sensor "restart" may be detected based at least in part on a signal from a temperature sensor in the sensor electronics package (e.g., the CGM transmitter). For example, if a subject intends to reuse a disposable sensor component, the subject will typically stop the sensor session and begin a new session without removing the transmitter from the disposable component. This "restart" scenario may be detected from the absence of a temperature signature associated with removal of the transmitter from the sensor.

When the transmitter is removed from the host and reconnected to a new sensor, if the sensor electronics are away from the host for a sufficient period of time (e.g., 1 minute), a temperature signature is observable that includes a temperature drop. FIG. 18A is a plot of temperature versus time where the sensor electronics package (Dexcom CGM transmitter) was removed from the sensor (Dexcom glucose sensor) for 1 minute at 1:27 PM. A temperature drop 1802 is visible in the temperature plot. FIG. 18B is a similar graph where the sensor electronics package was removed for 5 minutes at 3:34 PM. A larger temperature drop 1804 is visible in the temperature plot, and it takes more than 30 minutes for the sensor to return to the steady state temperature 1806 (approximately 33° C.) detected before the change.

In various examples, a disconnection event (e.g., removing the CGM transmitter from the sensor) may be identified based on the amount of temperature drop (e.g., 3°C or 5°C in a short period of time), the slope of the drop, or the consistency of the signal during the drop (lack of smoothness or variability), or a combination thereof.

A sensor restart may be identified from the absence of a disconnection event around the time of the session stop or start. In some examples, a disconnection event may be determined from a combination of a temperature signature (e.g., a temperature drop) and other information, such as a sensor session stopping. For example, when a temperature signature associated with a disconnection occurs immediately (or just before) a session ends, it may be inferred that the sensor electronics have been removed from the disposable sensor. When a sensor session is stopped and started but there is no temperature drop as described above and illustrated in FIGS. 18A and 18B, it may be inferred that the disposable sensor has been reused. This is because changing to a new sensor requires the sensor electronics (CGM transmitter) to be removed from the sensor. In some examples, sensor removal may be determined from a combination of a temperature signature and accelerometer data (e.g., rapid or large movements that may occur during transmitter disconnection followed by a temperature drop) or other sensor data.

13 is a flow chart diagram of an example method 1300 for determining that a continuous glucose (or other analyte) monitor has been restarted. The method 1300 may include, at operation 1302, receiving a temperature signal indicative of a temperature parameter from a temperature sensor on the continuous glucose monitor. The method 1300 may further include, at operation 1304, determining from the temperature signal that the continuous glucose monitor has been restarted. For example, as described above, a restart may be identified as an absence of a disconnection event in a temperature signature, optionally together with other sensor information. Upon detecting that the continuous glucose monitor has been restarted, a response action may be performed, for example, as described herein with respect to FIG. 34. For example, the sensor system may set a flag indicating that the continuous glucose monitor has been restarted. In some examples, upon detecting a session restart, the sensor system may terminate the restarted session.

Restarts can also be detected using a subcutaneous temperature sensor. When a temperature sensor is on the subcutaneous analyte sensor, the temperature reading from the sensor will typically be lower than body temperature (e.g., closer to ambient air temperature) when the sensor is first inserted, and the detected temperature can be expected to gradually rise to body temperature as the sensor absorbs heat from the body. In one example, determining from the temperature signal that the continuous glucose monitor has been restarted can include comparing a first temperature signal value before sensor initiation to a second temperature signal value after sensor initiation, and declaring the continuous glucose monitor to have been restarted when the comparison meets a similar condition. The similar condition can include a temperature range. For example, when a sensor is restarted (as opposed to being replaced), the temperature at the subcutaneous sensor typically does not change, or any change is gradual. When a sensor is replaced, a more significant temperature change can occur (e.g., the new sensor may indicate a different temperature than the old sensor).

In some examples, the temperature information may be used to determine the anatomical location or type of anatomical location where the sensor is worn. For example, the sensor may be worn on the arm or abdomen. The sensor (or sensor electronics) may experience a lower temperature when worn on the arm compared to the abdomen. This may be driven, for example, by the fact that the upper arm is farther away from the torso, or that the arm is more likely to be exposed to air, for example, when short-sleeve clothing is worn. Sensors worn on the arm may also experience greater fluctuations in temperature, especially during sleep (e.g., when the arm is more likely to be outside the sheets or blankets than the abdomen, at least for part of the night). In some examples, the anatomical location may be determined based on the average (e.g., mean or median) temperature over a specified period of time (e.g., during the first 24 hours of wearing). For example, the sensor device location may be declared to be on the abdomen when the average temperature meets a condition, such as when the average temperature exceeds a specified temperature threshold (e.g., 32°C). In another example, for example, an abdominal sensor location may be detected based on a fluctuation condition, e.g., a first standard deviation of temperature fluctuation during a specified period (e.g., nighttime or sleep period) being less than a specified amount (e.g., less than 1°C). In some examples, an abdominal location may be detected based on a combination of temperature and fluctuation conditions, e.g., an abdominal location may be declared when the average temperature exceeds a specified temperature threshold (e.g., 32°C) or when the first standard deviation of temperature fluctuation during a specified period is less than a specified amount (e.g., less than 1°C).

Figure 16 is a graph diagram showing plots of temperature (y-axis) versus time (x-axis) for two sensors. The first plot 1602 (dotted line) shows data from a sensor placed on the abdomen. The second plot 1604 (solid line) shows data from a sensor placed on the arm. During the first four hours, the sensor is not worn by the host (e.g., not yet inserted) and the data from the sensor is roughly correlated. After four hours, the sensor is inserted into the host and the temperature rises rapidly. After this transition, the fluctuation between the first plot 1602 and the second plot 1604 is evident, with the second plot 1604 (corresponding to the arm-worn sensor) showing lower temperatures and more intense fluctuations.

17 is a plot of standard deviation versus mean temperature over the first 24 hours for several dozen sensor devices. Using the method discussed above (SD>1.0 and mean temperature<32°C), we identified sensor devices located on the arms with high sensitivity (all but five arm-worn sensors were identified as such) and good specificity (only six abdomen-worn sensors were identified as being on the arms according to the method). In some examples, the exemplary temperature method may be combined with information from other sensors (e.g., accelerometer data) to further increase sensitivity and specificity. In some examples, a learned model (e.g., using a neural network) may be used to identify patterns or relationships, and the model may be applied to determine the location. Such an approach may achieve higher sensitivity or specificity. Although specific "arm" and "abdomen" locations are shown, other locations or classes may be used (e.g., waist location may be determined, or "torso" location may include both abdomen and waist).

14 is a flow chart diagram of an example method 1400 of determining an anatomical location of a sensor. The method may include, at operation 1402, receiving a temperature signal indicative of a temperature of a component of a continuous glucose sensor on a host. The method may include, at operation 1404, determining an anatomical location of the continuous glucose sensor on the host based at least in part on the received temperature signal. In some examples, the anatomical location may be determined based at least in part on a sensed temperature. In some examples, the anatomical location may be determined based at least in part on a variation in the temperature signal. For example, a greater temperature variation may be seen in a sensor inserted in a peripheral location (e.g., an arm) or in a location less likely to be covered by clothing than a sensor inserted on the abdomen or lower back.

In some examples, the method may further include receiving an accelerometer signal, and determining the anatomical location may include determining the anatomical location based on the accelerometer signal. For example, a higher activity level or more frequent postural changes (either of which may be determined from the accelerometer signal) may indicate a peripheral location (e.g., on the back of the arm), while a lower activity level or less frequent postural changes or more periodic postural changes (e.g., correlated with sleeping or sitting) may indicate an abdominal or lumbar location. In some examples, a distribution of the rate of change of the location may be used to identify the anatomical location. For example, a distribution skewed toward a higher rate of change may suggest a peripheral location (e.g., on the arm), and a distribution skewed toward a lower rate of change may suggest a location on the torso (e.g., on the abdomen). In another example, a neural network or other learned model may be used (e.g., using sensor data and, optionally, based on user-input data indicating a specified anatomical location) to learn patterns or relationships that may be used to determine or predict the anatomical location.

In operation 1406, in some examples, temperature compensation may be based at least in part on anatomical location. For example, the temperature compensation algorithm may take into account the fact that subcutaneous temperature in the abdomen or lower back may change more slowly than subcutaneous temperature in the arm, which may have a lower mass to act as a heat sink or heat source.

In some examples, compression may be detected based at least in part on a signal from a temperature sensor. Compression of the sensor may occur, for example, when a person lies or leans on the sensor, which may occur, for example, while sleeping. When the glucose sensor is compressed, a lower estimated glucose value may be generated. When the subject lies on the glucose sensor, the temperature of the sensor may increase. Compression of the sensor may be detected based at least in part on an increase in the temperature of the sensor. In one example, a rapid drop in glucose levels that coincides with or precedes an increase in temperature may indicate that the sensor is being compressed. In some examples, additional information such as activity information may be used along with the temperature. For example, a rapid drop in estimated glucose in combination with low activity (suggesting that the subject is not exercising) and an increase in sensor temperature (suggesting that the subject is lying on the sensor) may indicate a drop in compression. In some examples, an alert may be triggered in response to a possible drop in compression. For example, a notification may be delivered via the smart device or a sound may be emitted from the smart device or sensor, which may prompt the subject to move away from the sensor to allow an accurate estimated glucose value to be obtained.

In some examples, sleep may be detected based at least in part on temperature sensor information. For example, warmer temperatures may be observed during sleep. A more consistent temperature or temperature pattern may be observed during sleep. Sleep may be detected by applying a model or algorithm that detects periods of warm temperatures, consistent temperatures, or temperature patterns (e.g., binary patterns corresponding to covered or uncovered arm sensors), optionally along with other sensor information. In some examples, temperature information may be used along with posture information from a 3D accelerometer, activity information, respiration, or heart rate, or any combination thereof, to detect sleep. In some examples, alert behavior may be altered in response to sleep detection. For example, alert thresholds may be adjusted to reduce the number of alerts during sleep, or alert triggers may be adjusted to provide time to process hypoglycemic events, or only certain types of alerts (e.g., more urgent alerts) may generate audio when sleep is detected.

In some examples, compression detection, compensation, or alerts may be provided or modified during sleep. For example, when a person is sleeping and there is a sudden, rapid drop in the estimated glucose value, compression may be inferred based on the sleep condition or state and the sudden drop in the estimated glucose value, optionally in combination with other information, such as a discontinuity in the glucose curve, an increase in temperature, or other information.

In some examples, it may be desirable to reduce the cost of hardware included in an analyte sensor system, such as the analyte sensor system 8 of FIG. 1. For example, components of the analyte sensor system 8, such as the sensor electronics 12 and/or all or a portion of the continuous analyte sensor 10, may be disposable products that are used for a sensor session lasting several days and then discarded. Therefore, it may be desirable to obtain highly accurate temperature values from an inexpensive temperature sensor.

Various examples described herein are directed to systems and methods that utilize a trained temperature compensation model to generate compensated temperature values from a system temperature sensor. The trained temperature compensation model may, in some examples, compensate for factors that cause errors in the raw temperature data, such as noise or other nonlinearities. Utilizing a trained model to compensate temperature values from a system temperature sensor, as described herein, may generate acceptably accurate temperature values using less expensive or more readily available system temperature sensors. For example, using a trained model, as described herein, may, in some examples, enable the use of less expensive or more readily available temperature sensors, such as sensors included with or generated from suitable diodes in an application specific integrated circuit (ASIC) or other components of the analyte sensor system 8.

The temperature compensation model may be any suitable type of model, including, for example, a neural network, a state model, or any other suitable trained model. Inputs to the temperature compensation model may include, for example, raw temperature data and uncompensated temperature data. Raw temperature data may include data generated by a system temperature sensor to indicate temperature, such as, for example, current, voltage, counts, etc. Uncompensated temperature data may include data indicative of uncompensated temperature. For example, a temperature sensor may provide data indicative of temperature that, in some examples, is derived from the raw temperature data. In some examples, inputs to the temperature compensation model may include both raw temperature data and uncompensated temperature data. In some examples, outputs of the temperature compensation model may include compensated temperature values.

In some examples, the output of the temperature compensation model may include, in addition to or instead of a compensated temperature value, a sensor characteristic that describes a relationship between the raw temperature data generated by the system temperature sensor and a corresponding temperature value. For example, the output of the temperature compensation model may include a slope and an offset. The slope and offset may be applied to the raw temperature data generated by the system temperature sensor to generate a compensated temperature value.

The temperature compensation model may be trained, for example, utilizing a reference temperature sensor that is more accurate than the system temperature sensor. FIG. 24 is a flow chart diagram of an example method 2400 for training a temperature compensation model. The system temperature sensor and the reference temperature sensor may be positioned to measure the temperature of an object, such as, for example, a surface, a volume of liquid in a container, etc. In operation 2402, the object is heated and/or cooled to a first temperature. When the object is at various temperatures, inputs may be provided to the temperature compensation model in operation 2404. In response to the inputs, the temperature compensation model generates one or more model outputs. The one or more model outputs are compared to a reference temperature measured by a reference temperature sensor in operation 2406.

In operation 2408, the model parameters are modified based on the error between the reference temperature and the output of the temperature compensation model. The error is a model output and/or indicates the difference between a compensated temperature value generated using the model output and the reference temperature. The error is used to modify the parameters of the model. Method 2400 may be run and repeated, for example, until the model converges. The model may converge when the error of the temperature compensation model is consistently within an acceptable range. In some examples, a temperature compensation model is trained for each analyte sensor system 8. In other examples, the analyte sensor systems 8 and associated system temperature sensors may have similar characteristics that allow a temperature compensation model trained on one analyte sensor system 8 to be used on other analyte sensor systems 8, such as, for example, other analyte sensor systems 8 having similar components as the analyte sensor system 8 used to train the model, other analyte sensor systems 8 manufactured in the same batch as the analyte sensor system 8 used to train the model, etc.

FIG. 25 is a flow chart diagram of an example method 2500 for utilizing a trained temperature compensation model. In operation 2502, data is received from a system temperature sensor. The data may include, for example, raw temperature data and/or uncompensated temperature data. In operation 2504, the data received from the system temperature sensor is applied to a model to generate one or more model outputs. The model outputs may include a compensated temperature value and/or system temperature sensor parameters such as slope and offset that may be used to generate the compensated temperature value.

Optionally, at 2506, the model output is used to generate a compensated temperature value. For example, if the model output includes a compensated temperature value and/or if the model output does not include a system temperature sensor parameter, the compensated temperature value in operation 2506 may be omitted. At 2508, the compensated temperature value is applied. For example, the compensated temperature value may be applied to any of the methods described herein for utilizing temperature in conjunction with an analyte sensor.

In some examples, operations 2502 and 2504 are performed using a portion of the raw sensor data received from the sensor, for example, at the beginning of a sensor session. By applying the raw temperature values to a model, system temperature parameters such as slope and offset may be obtained. The system temperature parameters are applied to subsequently received raw sensor data to generate subsequent compensated temperature values.

As described herein, the exercise condition or state of the host can affect the temperature compensation model applied to generate the temperature-compensated glucose concentration. The exercise condition or state of the host can be determined in a variety of different ways, including, for example, utilizing a third sensor signal as described herein with respect to FIG. 7. In some examples, other techniques can be used to detect the exercise condition or state in addition to or instead of using a third sensor.

26 is a flow chart diagram of an exemplary method 2600 for detecting an exercise condition or state. The exemplary method 2600 detects an exercise condition or state of a host by examining the noise floor of a glucose sensor signal, the noise floor of a temperature parameter signal, or both. The noise floor is the level of noise associated with a signal. For example, the noise floor can be the sum of noise sources in a signal other than the value of interest. For example, the noise floor of a glucose sensor signal is the sum of noise sources in a signal other than glucose. The noise floor of a temperature parameter signal is the sum of noise sources in a temperature parameter other than an indication of temperature. In some examples, when a host is in an exercise condition or state, physiological behavior associated with exercise manifests itself in additional noise sources that affect the glucose sensor signal and/or the temperature parameter signal. Thus, the method 2600 detects the exercise condition or state by measuring the respective noise floors. The method 2600 can be implemented in an analyte sensor system 8, such as, for example, the sensor electronics 12 and/or the display devices 14, 16, 20.

The method 2600 may include, at operation 2602, accessing a glucose sensor signal. For example, the glucose sensor signal may be received from a continuous glucose monitor (CGM). The method 2600 may include, at operation 2604, accessing a temperature parameter signal. Accessing the temperature parameter signal may include, for example, receiving a signal indicative of a temperature, a temperature change, and/or a temperature offset.

The method 2600 may include, at operation 2606, determining a noise floor of the glucose sensor signal, the temperature parameter signal, or both. The noise floor or floors may be determined in any suitable manner. In some examples, the noise floor of a signal may be approximated by finding a minimum value of the signal. In another example, the noise floor of a signal may be determined using spectral analysis.

Method 2600 may include, at operation 2608, determining whether a noise floor threshold is met. In some examples, the threshold at operation 2608 is met if the noise floor of the glucose sensor signal is greater than a first threshold or if the noise floor of the temperature parameter signal is greater than a second threshold. In some examples, the threshold at operation 2608 is met if the noise floor of the glucose sensor signal is greater than a first threshold and the noise floor of the temperature parameter signal is greater than a second threshold.

If the noise floor threshold is met, the host is in an exercise condition or state. Thus, method 2600 includes modifying the temperature compensation based on the temperature parameter signal to account for the exercise condition or state at operation 2610. Examples of how this may be done are described herein with respect to methods 700 (e.g., 708 and 710) and method 800 (806). For example, as described herein, the temperature parameter signal may be applied to an exercise model before being used to generate a temperature-compensated glucose concentration. If the noise floor threshold is not met, the host may not be in an exercise condition or state, and an indication that the exercise condition or state is not present may be returned at operation 2612. Alternatively, instead of sending an indication that the exercise condition or state is not present, method 2600 may instead avoid modifying the temperature compensation at operation 2612.

Detecting that the host is in an exercise condition or state based on the rate of change of a temperature parameter as described herein. In some examples, this may be performed using a change distribution function. The change distribution function indicates the distribution of the rate of change over successive samples of a signal. FIG. 27 is a graph 2700 showing a first change distribution function 2702 indicative of a host in a resting (e.g., non-exercising) condition or state, and a second change distribution function 2704 indicative of a host in an exercise condition or state. In graph 2700, the horizontal axis indicates the rate of change of a temperature signal indicative of subcutaneous temperature at the glucose sensor. The vertical axis measures the cumulative distribution of the rate of change. As shown, the first change distribution function 2702 is approximately centered on a zero rate of change, meaning that approximately half of the rates of change between successive samples are greater than zero and approximately half are less than zero. The cumulative distribution function 2704 is skewed low, meaning that when the host is in an exercise condition or state, there are more rates of change less than zero than rates of change greater than zero. This can be exploited by examining the rate of change between temperature parameter signal samples, such as the example histogram 2706, as described herein.

28 is a flow chart diagram of an example method 2800 for detecting a motion condition or state using a distribution of rates of change in temperature parameter signal samples. Method 2800 may be performed in an analyte sensor system 8, such as, for example, sensor electronics 12 and/or display devices 14, 16, 20.

The method 2800 may include, at operation 2802, accessing a current temperature parameter signal sample. The method 2800 may include, at operation 2804, determining a rate of change between the current temperature parameter signal sample and a previous temperature parameter signal sample. The rate of change may be a difference. As described herein, the rate of change may be negative (e.g., if the current sample is less than the previous sample) or positive (e.g., if the current sample is greater than the previous sample). The method 2800 may include, at operation 2806, storing the current rate of change.

In operation 2808, a classifier is applied to the historical rates of change, including the newly stored rates of change. The classifier may, for example, represent a distribution of rates of change over a predetermined number of samples (e.g., 30 samples). The measured distribution of rates of change over the predetermined number of samples is compared to the classifier. In operation 2810, it is determined whether the measured distribution of rates of change satisfies the classifier. For example, the classifier may describe the number of measured rates of change between samples that fall into some range or a range of numbers of measured rates of change. An example of a classifier is shown in Table 4 below.

In the example of Table 4, the rate of change measure distribution satisfies the classifier if the number of measured rate of change below -0.4°C/min is greater than 2, the number of measured rate of change between -0.4°C/min and -0.2°C/min is greater than 10, and so on.

If the classifier is satisfied, the host is in an exercise condition or state. Thus, method 2800 includes modifying the temperature compensation based on the temperature parameter signal to account for the exercise condition or state at operation 2812. Examples of how this may be done are described herein with respect to methods 700 (e.g., 708 and 710) and method 800 (806). For example, as described herein, the temperature parameter signal may be applied to an exercise model before being used to generate a temperature-compensated glucose concentration. If the classifier is not satisfied, the host may not be in an exercise condition or state, and an indication of the absence of an exercise condition or state may be returned at operation 2814. Alternatively, instead of sending an indication of the absence of an exercise condition or state, method 2800 may avoid modifying the temperature compensation instead at 2814.

In some examples, any of the various temperature sensor configurations described herein may be used to measure the temperature of the analyte sensor during storage and/or shipping. For example, the peak temperature to which the analyte sensor is exposed prior to a sensor session may affect the performance of the sensor. For example, the peak temperature to which the analyte sensor is exposed prior to a sensor session may affect the initial sensitivity and/or baseline of the analyte sensor when inserted into the skin of a host. Also, in some examples, if the peak temperature to which the analyte sensor is exposed is too high, the sensor may no longer be suitable for use.

In various examples, an analyte sensor system, such as the analyte sensor system 8 of FIG. 1, is configured to periodically record temperature at the analyte sensor system during storage and/or transport using one or more of the various sensor arrangements described herein. FIG. 29 is a flow chart diagram of an exemplary method 2900 for recording temperature at the analyte sensor system during shipping. The analyte sensor system (e.g., its sensor electronics 12) can be programmed to perform the method 2900, for example, while the analyte sensor system is packaged for storage and/or transport.

The method 2900 may include, at operation 2902, the analyte sensor system waking up. For example, the processor of the analyte sensor system may be programmed to wake up periodically as described herein. Upon waking up, the analyte sensor system may measure the current temperature at operation 2904. The analyte sensor system may include any of the temperature sensor configurations described herein and may use one or more temperature sensor configurations to measure the temperature at operation 2904. The analyte sensor system may record the measured temperature at operation 2906. The measured temperature may be recorded, for example, in a data storage memory (e.g., 220 of FIG. 2) or another suitable data storage location in the analyte sensor system. At operation 2908, the analyte sensor system waits for one cycle. A cycle may be, for example, 10 minutes, one hour, one day, etc. After waiting one cycle, the analyte sensor system returns to operation 2902 and wakes up again as described.

The analyte sensor system may perform method 2900 while in storage and/or while being transported to a host for use. In this manner, the analyte sensor system may arrive at the host to begin a sensor session with a record of periodic temperature measurements stored in the data storage memory.

30 is a flow chart diagram of an exemplary method 3000 for initiating a sensor session having an analyte sensor session that includes recording periodic temperature measurements obtained from transport and/or storage of the analyte sensor system. In operation 3002, the analyte sensor system initiates a sensor session. This may occur, for example, when an analyte sensor of the analyte sensor system is positioned on a host, for example, when the analyte sensor is inserted into the skin of the host. In operation 3004, the analyte sensor system determines a peak temperature to which the analyte sensor system was exposed prior to the sensor session. This may include, for example, reading the record of periodic temperature measurements and determining a maximum temperature measurement from the record. The maximum temperature measurement may be a peak temperature measurement.

In operation 3006, the analyte sensor system determines whether the peak temperature measurement is greater than a threshold value. The threshold value may be, for example, the maximum temperature to which the analyte sensor can be exposed prior to a sensor session without impairing sensor performance during the session. If the peak temperature is greater than the threshold value, the analyte sensor system may abort the sensor session in operation 3010. In some examples, the analyte sensor system processes the determination that the peak temperature measurement is greater than the threshold value by performing an exception procedure 3400 described herein with respect to FIG. 34. This may include, for example, sending a message to one or more display devices, such as display devices 14, 16, 18, and/or 20, indicating that the analyte sensor or sensor system is not suitable for use and that a different sensor or analyte sensor system should be used.

If the peak temperature is not greater than the threshold in operation 3006, the analyte sensor system may select initial sensor session parameters based on the peak temperature in operation 3008. The initial sensor session parameters may be or may include, for example, initial sensitivity, initial baseline, etc. The initial sensor session parameters may be used by the analyte sensor system, for example, to generate an estimated analyte value using the raw sensor data. In some examples, the initial sensor session parameters are used after a sensor break-in period. In some examples, the analyte sensor applies a trained model to the peak temperature to determine the one or more initial sensor session parameters. In another example, a relationship between the peak temperature and the one or more initial sensor session parameters is stored in the analyte sensor system, for example, in a lookup table. In some examples, in addition to or instead of the peak temperature, the analyte sensor system may determine an average temperature, a median temperature, etc., or other suitable indicator of temperature during packaging.

In some examples, methods 2900 and/or 3000 include considering humidity in addition to or instead of temperature. For example, with reference to method 2900, the analyte sensor system may measure humidity upon wake-up. Humidity may be measured in any suitable manner. For example, systems and methods for measuring humidity in an analyte sensor based on the membrane impedance of the sensor are described in U.S. Patent Application No. 62/786,166, Attorney Docket No. 638PRV, filed December 28, 2018, entitled “ANALYTE SENSOR WITH IMPEDANCE DETERMINATION,” which is incorporated herein by reference in its entirety. With reference to method 3000, if the analyte sensor is exposed to a humidity outside of a predetermined range, the sensor session may be aborted. Additionally, the peak humidity to which the analyte sensor was exposed may be used to determine initial sensor session parameters.

In some examples, a diode may be used as a temperature sensor. A diode may be used as a temperature sensor, for example, by exploiting the temperature dependence of the voltage drop across the diode. Consider the equation for a Shockley diode, given by the following equation [10]:

In equation [10], V _T is given by the following equation [11].

In equations [10] and [11], I is the forward current through the diode. I _s is the reverse bias saturation current. V _D is the voltage across the diode. V _T is the thermal voltage, given by equation [11]. n is the ideality factor of the diode. k is the Boltzmann constant. q is the elementary electron charge. T is the absolute temperature of the diode in Kelvin. Rearranging equations [10] and [11] for voltage gives the following equation [12]:

To eliminate the unknown reverse bias saturation current, two known diode currents can be supplied to the diode, and the difference in voltage at the two different known currents is denoted as _ΔVD and is given by equation [13].

Solving for temperature, we get:

In some examples, the dependence of temperature T on the diode ideality factor n can be reduced by using a diode-connected NPN transistor with its base connected to its collector (e.g., "diode-connected") as the diode. In this configuration, the ideality factor n approaches 1 and can be eliminated from equation [14].

Various examples utilize the relationship of equation [14] to measure temperature in an analyte sensor system using a diode. For example, the NPN transistor and associated circuitry described herein may be less expensive than a suitably accurate temperature sensor, and in some examples may be much less expensive.

FIG. 31 is a diagram of an example circuit configuration 3100 that can be implemented in an analyte sensor system to measure temperature using a diode. The circuit configuration 3100 includes first and second current sources 3102, 3104 and a diode-connected NPN transistor 3106. Although a diode-connected transistor 3106 is shown in FIG. 31, in other examples, a different type of diode may be used.

Current source 3102 is a constant current source that provides a current of approximately 10 μA in this example. Current source 3104 is a pulsed current source that provides 40 μA pulses. Current sources 3102, 3104 may be implemented in any suitable manner, for example, utilizing one or more transistors. Current from current source 3102 and current source 3104 is provided to diode-connected transistor 3106 such that the current in diode-connected transistor 3106 is the sum of the current from current source 3102 and the current from current source 3104. This is illustrated by plot 3108. In this example, when pulsed current source 3104 is on, the current provided to diode-connected transistor 3106 is 50 uA, the sum of the constant 10 uA from current source 3102 and the pulsed 40 uA from current source 3104. When the pulsed current source 3104 is off, the current supplied to the diode-connected transistor 3106 is 10 uA provided by the constant current source 3102. The current sources 3102, 3104 provide current through resistors 3110, 3112 in this example.

In this configuration, as shown by plot 3108, diode-connected transistor 3106 receives two known currents. As demonstrated herein, the difference between the value of the voltage drop across diode-connected transistor 3106 at a first current (V1) and the value of the voltage drop across diode-connected transistor 3106 at a second current (V2) indicates the temperature of the pn junction in diode-connected transistor 3106.

To measure the voltage difference, the sample and hold (S/H) circuit 3116 receives at its input a voltage value indicative of the voltage drop across the diode-connected transistor 3106. At its clock input, the S/H circuit 3116 receives an indication of when the pulsed current source 3104 is off. This may be accomplished, for example, by inverting the signal generated by the pulsed current source 3104 using an inverter 3118. As a result, the output of the S/H circuit 3116 may be a voltage value V1 indicative of the voltage drop across the diode-connected transistor 3106 at the current provided by the first current source 3102.

The dual slope integrating analog-to-digital converter (ADC) 3114 may be used to convert the difference between the voltage value V1 and the voltage value V2 into a digital signal that may be consumed, for example, by a processor of the analyte sensor system. The dual slope integrating ADC 3114 comprises a first input 3120 and a second input 3122. The comparator 3124 has a non-inverting input that is grounded and an inverting input that is coupled to a switch 3128. The switch 3128 alternately connects the first input 3120 (through resistor RA) or the second input 3122 (through resistor RB) to the inverting input. The capacitor 3126 is coupled between the inverting input of the comparator 3124 and the output VOUT of the ADC 3114.

In the example of FIG. 31, the output of S/H circuit 3116, representing V1, is provided to input 3122 of ADC 3114. The voltage drop across diode-connected transistor 3106 is provided to input 3120. Switch 3128 is clocked to provide input 3120 to the inverting input of comparator 3124 when pulsed current source 3104 is on, and to provide input 3122 to the inverting input when current source 3104 is off.

Thus, when current source 3104 is off, capacitor 3126 is charged to a voltage V1, which is the voltage drop across diode-connected transistor 3106 from current source 3102. When current source 3104 is on, switch 3128 connects input 3120 to the inverting input, causing capacitor 3126 to charge to voltage V2. Voltage V2 is the voltage drop across diode-connected transistor 3106 due to the combined currents of current sources 3102, 3104. When current source 3104 is off again, switch 3128 connects voltage V1, and the voltage at capacitor 3126 (and VOUT) decays to V1. This is shown by plot 3130, which shows VOUT on the vertical axis and time on the horizontal axis. When VOUT is increasing, current source 3104 is on and switch 3128 connects V2 to the inverting input. When VOUT is decaying, current source 3104 is off and switch 3128 connects V1 to the inverting input. The time it takes for VOUT to decay from V2 to V1 indicates the difference between V2 and V1. This can be used to derive the temperature at diode-connected transistor 3106, for example, according to Equation 6 above.

In some examples, the circuit configuration 3100 includes a comparator 3132 that compares the output of the S/H circuit 3116, which indicates a voltage value V1, with the VOUT output of the ADC 3114. The output of the comparator (COMP OUT) may change state when VOUT is less than or equal to V1. Thus, the sensor electronics of the analyte sensor system may measure the difference between V1 and V2 by starting a digital counter when the switch 3128 connects to the input 3122 and stopping the digital counter when the comparator output COMP OUT changes state.

In some examples, an AND gate 3134 is provided to generate a logical AND of the comparator output (COMP OUT) and a clock signal. The output of AND gate 3134 may be used to stop a digital counter. This may ensure that the comparator changes state only when the voltage on capacitor 3126 is decaying.

FIG. 32 is a flow chart diagram of a method 3200 for measuring temperature in an analyte sensor system using a diode, such as the diode-connected transistor 3106 of FIG. 31. The method 3200 may include, in operation 3202, applying a first current across the diode. The method 3200 may also include, in operation 3204, measuring a voltage V1 indicative of a voltage drop across the diode at the first current. The method 3200 may also include, in operation 3206, applying a second current to the diode and, in operation 3208, measuring a second voltage V2 indicative of a voltage drop across the diode at the second current.

The first voltage V1 and the second voltage V2 are provided to a dual slope integrating ADC in operation 3210. In operation 3212, the time for the output of the ADC to decay from the second voltage V2 to the first voltage V1 is measured using, for example, a digital timer. The result may be a digital value indicative of the temperature at the diode, as described herein.

One way in which temperature can affect the performance of an analyte sensor system, such as a glucose sensor system, as described herein, relates to temperature-dependent compartment bias. A glucose sensor is inserted into the host's skin at an insertion site. Under the host's skin, the glucose sensor directly measures the glucose concentration at the insertion site, e.g., in the interstitial fluid present at the insertion site. However, the glucose concentration in the interstitial fluid may not be the same as the glucose concentration in the patient's blood. Compartment bias refers to the difference between the glucose concentration at the insertion site of the glucose sensor and the glucose concentration in the host's blood.

Compartment bias can arise due to glucose consumption by cells in the host. For example, glucose from the host's bloodstream is provided to host cells in the capillaries of the host's vasculature. Glucose diffuses from the capillaries to the host's cells. Cells between the nearest capillary or capillary system and the insertion site consume glucose. Because of this consumption, the glucose concentration at the insertion site (also called the interstitial fluid glucose concentration) is lower than the blood glucose concentration (also called the blood glucose concentration or capillary glucose concentration). The amount by which the interstitial fluid glucose concentration is below the blood glucose concentration is a compartment bias.

In some examples, the rate at which glucose diffuses from the host's capillaries to the insertion site and/or the rate at which glucose is consumed by cells between the host's capillaries and the insertion site varies with temperature. For example, the warmer the host's skin is, the faster glucose may diffuse. As a result, the glucose sensor system may apply a compartment model that compensates the glucose sensor signal for compartment biases that may be temperature dependent. An exemplary compartment model is given by Equation [15].

In equation [15], IG(t) is the glucose concentration in the interstitial fluid.

is the first derivative of the interstitial fluid glucose concentration, I G (t), over time. BG (t) is blood glucose. The values τ ₁ and τ ₂ are model time parameters. Equation [15] is a differential equation that can be solved to derive the model relationship between the interstitial fluid glucose concentration, I G , and the blood glucose concentration, B G , as given by equation [16].

The time parameters τ ₁ and τ ₂ may be dependent on temperature. For example, the glucose sensor system may model the time parameters τ ₁ and τ ₂ as a function of temperature. When the glucose sensor system receives the glucose sensor signal and the temperature sensor signal, the glucose sensor system may utilize the temperature sensor signal to derive the time parameters τ ₁ and τ ₂ and then determine the compensated glucose concentration using the time parameters in the compartment model, such as the time parameters given by equations [15] and [16].

In some examples, the glucose sensor system utilizes a compartment model that includes a single time parameter τ that is applied to both the interstitial fluid glucose concentration term I and the blood glucose concentration term B. In some examples, the difference between the time parameters τ ₁ and τ ₂ of the compartment model of Equations [15] and [16] is related to glucose consumption of cells between the host capillaries and the insertion site. Thus, in some examples, a single time parameter τ may be used by taking into account glucose consumption. An exemplary compartment model that takes into account glucose consumption is given by Equation [17]:

In equation [17], C(t) is a consumption term.

The consumption term C(t) can, in some examples, be modeled as given by equation [18].

In equation [18], V _max is the maximum consumption rate of the host cells. K _m is the glucose concentration at which the consumption rate is half V _max . [s _i ] is the cell layer glucose concentration at the i th cell layer between the host capillary and the insertion site. As shown in equation [18], the consumption rate is obtained by summing the number of cells n per unit volume (e.g., per dL).

Figure 33 illustrates an exemplary sensor insertion site 3300 showing the cell layers between the sensor insertion site 3300 and the host capillary site. In this example, five cell layers are shown for i=0-4. Cells in layer 0, such as cell 3302, have a cell layer glucose concentration S0. Cells in layer 1, such as exemplary cell 3304, have a cell layer glucose concentration S1. Cells in layer 2, such as exemplary cell 3306, have a cell layer glucose concentration S2. Cells in layer 3, such as exemplary cell 3308, have a cell layer glucose concentration S3. Cells in layer 4, such as exemplary cell 3310, have a cell layer glucose concentration S4.

In some examples, the glucose sensor system may apply equations [17] and [18] assuming that the cell layer glucose concentration [si] is constant regardless of distance from the sensor insertion site. For example, in some examples, the cell layer glucose concentration [si] for all cells is assumed to be the average of the interstitial fluid glucose concentration IG and the blood glucose concentration. Using this assumption, equations [17] and [18] may be approximated as given by equation [19].

Solving equation [19] for the blood glucose concentration results in a model that can be used to generate a compensated glucose concentration. For example, a glucose sensor system may utilize a temperature sensor signal to determine a value of τ and then apply τ to the solution of equation [19] to generate a compensated glucose concentration.

In another example, the glucose sensor system may apply equations [17] and [18], assuming that the consumption term from the sum in equation [18] is a linear function of i, for example, as illustrated by equation [20].

In equation [20], a is the slope and b is the offset. Given this assumption, the consumption given by equation [18] above reduces to the form shown by equation [21].

The compartment bias model is given by equation [22]:

Solving equation [22] for the blood glucose concentration results in a model that can be used to generate a compensated glucose concentration. For example, a glucose sensor system may utilize a temperature sensor signal to determine a value of τ and then apply τ to the solution of equation [22] to generate a compensated glucose concentration.

FIG. 34 illustrates an example of a process flow 3400 that may be performed in an analyte sensor system that uses temperature compensation to respond to an exception. In operation 3402, the analyte sensor system determines whether an exception has occurred. An exception may occur, for example, when a state and/or operation of the analyte sensor system deviates from an expected or preferred status or operation, for example, as described herein. In some examples, the analyte sensor system sets a flag, bit, variable, or other value (e.g., stored in memory) when it encounters an exception. The flag, bit, variable, or other value may be a register location in the processor or other memory location. The analyte sensor system may set a flag, bit, variable, or other value by writing a particular value to an associated memory location. Detecting the exception in operation 3402 may include determining whether one or more exception flags are set. For example, when the analyte sensor system encounters a glucose rate of change that exceeds a specified value, the analyte sensor system may set a flag indicating a glucose rate of change exception. In another example, if a comparison between a conductance-compensated temperature estimate and a conductance-free subcutaneous temperature estimate shows a difference greater than a set threshold, the analyte sensor system may set a flag.

If an exception is detected in act 3402, the analyte sensor system selects a response action in act 3403. The analyte sensor system may select a response action based on the exception that was present. For example, as described herein, different exceptions may be indicated by setting different flags, bits, variables, or other values in a processor or memory location associated with the analyte sensor system. Determining the response action may include determining which exception flag, bit, variable, or other value was set to trigger the exception in act 3402 and selecting one or more corresponding response actions 3404, 3406, 3408, 3410.

For example, if the exception indicates that the analyte sensor system session should not continue, the analyte sensor system may perform a response action in operation 3404 that includes terminating the current sensor session. The response action may include terminating the current sensor session if the exception occurred because, for example, a continuous glucose sensor was restarted, as described herein with respect to FIG. 13. In another example, the response action may include terminating the current sensor session if the peak temperature during the session exceeds a threshold value, as described herein with respect to FIG. 30.

In some examples, the response action is or includes pausing temperature compensation of the estimated analyte value generated by the analyte sensor system in operation 3406. For example, if the exception is triggered when the glucose rate of change of the continuous glucose sensor system exceeds a specified value, the response action may include pausing temperature compensation in operation 3406. Also, if the exception occurs when the confidence in the temperature compensated analyte value decreases, for example, as described herein with respect to FIGS. 15A-15D, the response action may include pausing temperature compensation in operation 3406.

In some examples, the response action includes modifying the temperature model being used to determine the in vivo sensor temperature from the ex vivo temperature in act 3408. Modifying the temperature model may include, for example, selecting and/or modifying one or more model parameters, such as the slope and/or offset of a linear model. The response action to the exception may include modifying the temperature model in act 3408 if the exception is triggered by a condition or state of the in vivo sensor temperature, such as, for example, when the sensor temperature exceeds a threshold, when the sensor temperature rate of change exceeds a threshold rate of change, and/or when the residual temperature exceeds a threshold. Modifying the temperature model may also include selecting a different model, for example, as described in more detail with respect to FIG. 40 herein. In some examples, modifying the temperature model includes using a constant value for the in vivo sensor temperature. For example, the in vivo sensor temperature may be limited such that the detected temperature is greater than a threshold, and the threshold (or another suitable value) is used as an input for temperature compensation instead of the measured value.

In some examples, the response action includes pausing or preventing display of the estimated analyte value determined by the analyte sensor in operation 3410. For example, the response action 3410 may be selected if the exception indicates that the accuracy of the estimated analyte value produced by the analyte sensor system is likely to be temporarily affected. In some examples, the response action 3410 may be selected when the in vivo sensor temperature and/or temperature gradient exceeds a threshold value.

When the response action is or includes terminating the current sensor session in operation 3404, the analyte sensor system may not begin measuring, storing, and/or displaying the estimated analyte value again until a new sensor session is initiated, e.g., upon subcutaneous or other in vivo insertion of a new sensor. When the response action included pausing temperature compensation in operation 3406, modifying temperature compensation in operation 3408, and/or pausing display of the estimated analyte value in operation 3410, the analyte sensor system may periodically re-execute process flow 3400, e.g., by determining whether the exception still exists in operation 3402.

If no exception exists at operation 3402, the analyte sensor system may wait until the next time process flow 3400 is executed and re-execute operation 3402. In some examples, process flow 3400 is executed periodically (e.g., every 30 seconds, every 5 minutes, every 10 minutes, etc.). In some examples, exception process 3400 is executed each time an estimated analyte value is generated from the sensor signal. Also, in some examples, exception process 3400 is executed each time an estimated analyte value is output to a display, such as one of display devices 14, 16, 20.

In optional operation 3412, the analyte sensor system may modify a previously performed response action. For example, if the previous response action included pausing the display of the estimated analyte value in operation 3410, the analyte sensor system may determine whether the conditions prompting the response action in operation 3410 are still met. If the conditions are no longer met, the analyte sensor system may begin to display the estimated analyte value result again. In another example, if the previous response action included pausing temperature compensation in operation 3406, the analyte sensor system may determine whether the conditions prompting the response action in operation 3406 are still met. If the conditions are no longer met, the analyte sensor system may begin performing temperature compensation again. Similarly, in some examples, if the previous response action included modifying the temperature compensation model in operation 3408, the analyte sensor system may determine whether the conditions prompting the response action in operation 3408 are no longer met or the system has otherwise changed. If conditions or conditions of the sensor, host, etc. have changed and additional modifications to the temperature compensation model are required, the modifications may be made. The modification to the temperature compensation model may be a reversion of the temperature compensation model to a previous form and/or a different modification to the temperature compensation model.

As described herein, the temperature at a subcutaneous sensor or other in vivo sensor may be determined using a temperature at an in vitro temperature sensor location, such as a CGM transmitter or other in vitro sensor hardware. FIG. 35 is a block diagram illustrating a temperature model 3500. In this example, the temperature model 3500 receives as an input an in vitro temperature measured by a temperature sensor positioned ex vivo. In various examples herein, the in vitro temperature or other non-subcutaneous temperature is denoted as T _Tx . The in vitro temperature T _Tx is obtained in some examples by a temperature sensor at a transmitter, such as one or more of temperature sensors 240, 242, 268, 270, etc. However, it should be understood that the various examples described herein may also be performed with an in vitro temperature measured at an in vitro location other than the transmitter of the analyte sensor system.

Model 3500 operates on the in vitro temperature T _Tx to generate a subcutaneous or other in vivo sensor temperature. The in vivo sensor temperature is shown in FIG. 35 as T _WE or working electrode temperature. However, in various examples, the in vivo sensor temperature may be obtained at other in vivo portions of the sensor, including, for example, a reference electrode, a counter electrode, or other in vivo portions of the analyte sensor.

As described herein, model 3500 may take a variety of different forms. In some examples, model 3500 may be a PDE model based on Penne's bioheat equation given by equation [7] above. Also, in some examples, model 3500 may be a linear approximation of the relationship between the in-vivo sensor temperature _TWE and the in-vivo temperature _TTX , as described herein with respect to equation [6].

In some examples, model 3500 includes a linear time-invariant (LTI) approximation of Penne's bioheat equation. For example, an LTI approximation of Penne's bioheat equation may be more accurate than the linear approximation given by equation [6], but requires fewer processing resources than solving Penne's bioheat equation for each desired in vivo analyte sensor temperature. An exemplary LTI approximation of Penne's bioheat equation is given by equation [23] below:

In Equation 23, Y is a 1×n vector containing the in-vivo sensor temperatures derived over time. _{T x Toeplitz} is a Toeplitz matrix created from the normalized values of the in-vivo temperatures. The vector

Multiplying the vector by T _{x Toeplitz} gives an approximation of the internal body temperature that would be returned by the Penne PDE bioheat equation. To apply the LTI approximation, the vector

The value of T can be determined utilizing the in vitro temperature _T and the value of the sensor temperature _T determined using the Penne PDE. For example, the value of the in vitro temperature _T can be normalized as given by equation [23] below and convolved with T given by equation [24] below.

In formula [23],

is the average of the T _Tx values over a given period of time. L is the length of the T _Tx matrix (e.g., the number of ex vivo temperature values used). The value of the in vivo sensor temperature T _WE may be determined using Penne's PDE and normalized, for example, as given by equation [25].

In formula [25]

is the average in vivo sensor temperature determined using Penne's equation over a given period of time.

may be determined, for example, as given by equation [26].

This result can be fitted to the parametric equation given by equation [27].

The parameters _c1 , _c2 , and _c3 in equation [27] can be used as parameters of an LTI approximation that is used to generate a value of the in-vivo sensor temperature without solving the Penne bioheat equation. For example, the in-vivo sensor temperature _TWE is calculated using the measured in-vitro temperature _TTX and the vector given by equation [28]:

can be found using a parametric approximation of

The LTI approximation or other linear approximation of the relationship between the in vitro temperature T _Tx and the in vivo sensor temperature T _WE may be accurate over the range in which the assumptions of the LTI approximation or other linear approximation are true. For example, the LTI approximation or other linear approximation may be accurate over the range in which the in vitro temperature T _Tx and the in vivo sensor temperature T _WE have a linear relationship. In addition, the LTI approximation is accurate over the range in which the relationship between the in vitro temperature T _Tx and the in vivo sensor temperature T _WE is linear and time invariant.

Over a range of in vitro temperatures T _Tx and in vivo sensor temperatures T _WE where the model assumptions do not hold, the accuracy of the model may be compromised, and in some instances may be significantly compromised. Thus, an exemplary analyte sensor system may be configured to determine an exception when the in vitro temperature T _Tx and/or the in vivo sensor temperature T _WE are outside the range of the model assumptions.

In various examples, the assumptions of the model 3500 based on linear approximation may not hold when the in-vivo sensor temperature _TWE is outside of a given range. Thus, some examples of the analyte sensor system described herein detect an exception when the in-vivo sensor temperature _TWE is outside of a given range. FIG. 36 shows a chart 3600 illustrating an example relationship between the in-vitro temperature _TTX and the in-vivo sensor temperature _TWE , indicating the range of the in-vivo sensor temperature _TWE that will trigger an exception. In the chart 3600, the horizontal axis indicates the in-vivo sensor temperature _TWE and the vertical axis indicates the in-vivo temperature _TTX . The line 3602 represents an example linear relationship between the in-vitro temperature _TTX and the in-vivo sensor temperature _TWE . For example, the in-vitro temperature 3608 may correspond to the in-vivo sensor temperature 3610, as shown by the line 3602.

36, the assumptions of the linear model shown by line 3602 may not hold when the in vivo sensor temperature _TWE is below the lower temperature limit 3612 or above the upper temperature limit 3614. For example, the relationship between the in vivo sensor temperature TWE and the in vitro temperature TTX. in region 3604 when the in vivo sensor temperature _TWE is below the lower temperature limit 3612, and the relationship between the in vivo sensor temperature _TWE and the in vitro temperature _{TTX .} in region 3606 when the in vivo sensor temperature _TWE is above the upper temperature limit _3614. The values of the upper temperature limit 3612 and the lower temperature limit 3614 may vary based on various factors, such as, for example, the type of model used, the parameters of the model used, the characteristics of the host, the characteristics of the sensor, etc. In some examples, the upper temperature limit 3614 may be between 40°C and 60°C. In some examples, the upper temperature limit 3614 is about 45°C. In some examples, the lower temperature limit 3612 is between about 10° C. and 30° C. In some examples, the lower temperature limit is about 25° C.

In some examples, an analyte sensor having the characteristics illustrated by chart 3600 may implement limits or guardrails for the in-vivo sensor temperature _TWE . For example, if application of the model results in an in-vivo sensor temperature _TWE , _TWE , that is lower than the lower temperature limit 3612 or higher than the upper temperature limit 3614, the analyte sensor system may detect the exception and prompt appropriate responsive action, for example, by setting an appropriate flag in the analyte sensor system's processor or memory. In some embodiments, one or more limits or guardrails may be applied to the in-vivo temperature _TTX instead of or in addition to the in-vivo sensor temperature _TWE , for example, as discussed in FIG. 37. Response actions to an out-of-range in-vivo sensor temperature _TWE may include, for example, pausing temperature compensation, modifying the temperature model being used to convert the in-vivo temperature _TTX to a corresponding in-vivo sensor temperature _TWE , pausing display of the determined estimated analyte value, terminating the current sensor session, etc.

In some examples, the analyte sensor system may be configured to limit the in vitro temperature T _Tx to a range of in vitro temperatures. For example, the assumptions of the model used to convert the in vitro temperature T _Tx to an in vivo sensor temperature T _WE may not hold outside the range of in vitro temperatures T _Tx . FIG. 37 shows a version of the chart 3600′ of FIG. 36 including regions 3704, 3702 corresponding to situations where the in vitro temperature T _Tx is above an upper temperature limit 3708 and/or below a lower temperature limit 3706. In some examples, the temperature conversion from the in vitro temperature T _Tx to a corresponding in vivo sensor temperature T _WE may not be performed when the in vitro temperature T _Tx is above an upper temperature limit 3708 and/or below a lower temperature limit 3706. In some examples, the analyte sensor system sets an exception when the measured in vitro temperature T _Tx is above an upper temperature limit 3708 and/or below a lower temperature limit 3706. The exception may be handled, for example, as described with respect to FIG. 34.

In some examples, the upper temperature limit 3708 may be between about 60 C and about 70 C. In some examples, the upper temperature limit 3708 may be about 65 C. In some examples, the lower temperature limit 3706 may be between about 5 C and -5 C. In some examples, the lower temperature limit 3706 may be about 0 C. The in vitro temperature range indicated by the upper temperature limit 3708 and the lower temperature limit 3706 may be selected to exclude from consideration values of the in vitro temperature T _Tx that are not related to the in vivo sensor temperature T _WE according to the assumptions of the model based on linear approximation. Also, in some examples, the upper and lower temperature limits 3708, 3706 may be selected based on the capabilities of the temperature sensor 240, 242, 268, 270, etc. used to measure the in vitro temperature T _Tx . For example, the accuracy of the selected temperature sensor may fall outside the indicated range, causing the derived in vivo sensor temperature T _WE to be inaccurate as well.

38 is a flow chart illustrating an example of a process flow 3800 that may be performed by an analyte sensor system to determine an in vivo sensor temperature from an in vitro temperature measurement taking into account model parameters. The process flow 3800 illustrates an example technique for restricting the temperature conversion to both a range of in vitro temperatures T _Tx and a range of in vivo sensor temperatures T _WE .

In operation 3802, the analyte sensor system may access an in vitro temperature sensor signal. The in vitro temperature sensor signal may have been captured by an in vitro temperature sensor, such as, for example, one or more of the temperature sensors 240, 242, 268, 270 described herein. In operation 3806, the analyte sensor system determines whether the in vitro temperature signal indicates a temperature outside an in vitro temperature range. The in vitro temperature range may correspond, for example, to a range between a lower temperature limit 3706 and an upper temperature limit 3708 shown in FIG. 37. If the in vitro temperature T _Tx is outside the in vitro temperature range, the accessed in vitro temperature sensor signal may be discarded in operation 3806. In some examples, operations 3802 and 3804 may be implemented by firmware of the in vitro temperature sensor 240, 242, 268, 270, etc. For example, the in vitro temperature sensor 240, 242, 268, 270 may include firmware executable to interrogate the sensor's own signal. If the sensor signal indicates a temperature outside the in vitro temperature range, the sensor may discard the signal, for example, by not providing a signal to other components of the analyte sensor system.

If the in vitro temperature sensor signal indicates an in vitro temperature T _Tx that is not outside the in vitro temperature range, the analyte sensor system may apply the temperature model to the in vitro temperature to generate a corresponding in vivo sensor temperature T _WE in operation 3808. In operation 3810, the analyte sensor system may determine whether the determined in vivo sensor temperature T _WE indicates an exception. The in vivo temperature T _WE may indicate an exception, for example, if it is outside the in vivo temperature range. Also, as described herein, the in vivo temperature T _WE may indicate an exception if the rate of change of the in vivo temperature T _WE is greater than a threshold and/or if the residual temperature is outside the residual temperature range. Additional exemplary details regarding determining whether the determined in vivo sensor temperature T _WE indicates an exception are provided herein with respect to FIG. 39. If an exception is detected in operation 3810, the analyte sensor system may indicate the exception in operation 3812, for example, by setting an appropriate exception flag. The exception may be addressed by taking responsive actions, for example, as described herein with respect to FIGURE 34. If the in-vivo sensor temperature _TWE does not indicate an exception, the analyte sensor system may generate a temperature compensated analyte value in operation 3814, for example, as described herein.

39 is a flow chart 3900 illustrating an example of detecting an exception based on a determined in-vivo sensor temperature _TWE . The process flow 3900 has three branches 3902, 3904, 3906 that test for different temperature conditions that may cause an exception. Branch 3902 tests for an exception due to the in-vivo sensor temperature _TWE being outside of the in-vivo temperature range. Branch 3904 tests whether the rate of change of the in-vivo sensor temperature _TWE exceeds a rate of change threshold. Branch 3906 tests for a temperature residual exception when the residual temperature (e.g., the difference between the in-vivo temperature _TTX and the in-vivo sensor temperature _TWE ) exceeds a temperature residual threshold. In some examples, one or more of the branches 3902, 3904, 3906 may be omitted.

Referring to branch 3902, initially, the analyte sensor system tests the determined in-vivo sensor temperature against one or more temperature conditions related to the in-vivo sensor temperature range in operation 3908. For example, if the in-vivo sensor temperature is within the in-vivo temperature range, the in-vivo sensor temperature may satisfy the temperature condition. If the in-vivo sensor temperature does not satisfy the condition (e.g., if it is outside the in-vivo temperature range), the analyte sensor sets an in-vivo temperature exception in operation 3912 and proceeds to detect exception operation 3910. Setting the in-vivo temperature exception may include, for example, setting a flag in a processor and/or memory of the analyte sensor system, where the flag corresponds to the out-of-range in-vivo sensor temperature exception. If the in-vivo sensor temperature satisfies the condition (e.g., not outside the in-vivo temperature range), the analyte sensor may proceed to detect exception operation 3910 without setting the in-vivo temperature exception.

Referring now to branch 3904, in operation 3914, the analyte sensor system may determine a rate of change of the in-vivo sensor temperature _TWE . This may include, for example, examining the current in-vivo sensor temperature _TWE as well as one or more previously measured in-vivo sensor temperatures _TWE . In operation 3916, the analyte sensor system may determine whether the rate of change of the in-vivo sensor temperature satisfies a temperature condition related to a change range threshold. The determined rate of change may satisfy the condition if it is less than the rate of change threshold and may not satisfy the threshold if it is equal to or greater than the threshold. If the rate of change does not satisfy the rate of change temperature condition, the analyte sensor system may set an in-vivo temperature rate of change exception in operation 3918 and proceed to exception detection operation 3910. Setting the in-vivo temperature rate of change exception may include, for example, setting a flag in a processor and/or memory of the analyte sensor system, the flag corresponding to the out-of-range in-vivo sensor temperature rate of change exception. If the rate of change meets the rate of change temperature condition (eg, does not exceed a threshold), the analyte sensor system may proceed to detect exception operation 3910 without setting a biotransformation rate exception.

Now referring to branch 3906, the analyte sensor system determines a residual temperature in operation 3920. The residual temperature may be the difference between the in vitro temperature T _Tx and the in vivo sensor temperature T _WE . Thus, this operation 3920 may utilize both the in vitro temperature T _Tx and the in vivo sensor temperature T _WE . In operation 3922, the analyte sensor system determines whether the residual temperature determined in operation 3920 satisfies a temperature condition related to a residual temperature threshold. The residual temperature may satisfy the condition if it is within a determined residual temperature range or is less than the residual temperature threshold. The residual temperature may not satisfy the condition if it is outside the determined range or is greater than the threshold.

If the residual temperature does not meet the temperature condition, the analyte sensor system may set a residual temperature exception in operation 3924 and proceed to exception detection operation 3910. Setting the residual temperature exception may include, for example, setting a flag in a processor and/or memory of the analyte sensor system, where the flag corresponds to the out-of-range residual temperature exception. If the residual temperature meets the temperature condition, the analyte sensor system may proceed to exception detection operation 3910 without setting a residual temperature exception.

In a detect exception operation 3910, the analyte sensor system may determine whether any of the exceptions tested by branches 3902, 3904, 3906 are set. If any of the exceptions are set, the analyte sensor system may perform a response action, for example, as described herein with respect to FIG. 34.

In various examples, the assumptions of the temperature model for relating the in vitro temperature to the in vivo sensor temperature are different across different ranges of in vivo sensor temperatures and/or in vivo temperatures. Thus, in some examples, the analyte sensor system may use different models across different ranges of in vitro and/or in vivo temperatures. Using different models may mean, for example, using different models having different forms and/or using the same model form with different parameters for different ranges of in vitro and/or in vivo temperatures.

40 shows a chart 4000 illustrating an example relationship between the in-vitro temperature T _Tx and the in-vivo sensor temperature T _WE represented by lines 4002, 4004, 4006, 4008. Each line 4002, 4004, 4006, 4008 shows a temperature model that may be used over a different range of the in-vivo sensor temperature T _WE .

In the chart 4000, the horizontal axis indicates the in-vivo sensor temperature _TWE and the vertical axis represents the in-vivo temperature _TTX . In the example of Fig. 40, line 4002 represents a linear approximation of the relationship between the relationship between the in-vivo temperature _TTX and the in-vivo sensor temperature _TWE between the in-vivo sensor temperatures 4012 and 4014. Line 4004 represents a linear approximation of the relationship between the relationship between the in-vivo temperature _TTX and the in-vivo sensor temperature _TWE between the in-vivo sensor temperature 4014 and the in-vivo sensor temperature 4016. Line 4006 represents a linear approximation of the relationship between the relationship between the in-vivo temperature _TTX and the in-vivo sensor temperature _{TWE between the in-vivo sensor temperature 4016 and the in-vivo sensor temperature 4018. Line 4008 represents a limited in-vivo sensor temperature 4018 that may be applied when the in-vivo sensor temperature TWE is} particularly greater than the in-vivo sensor temperature 4018. For example, the in vivo sensor temperature 4108 may be a default temperature to be used above and including the in vivo sensor temperature TWE region 4010 that represents an in vivo sensor temperature _TWE below the in vivo sensor temperature 4012 may not be used.

Lines 4002, 4004, 4006, 4008 may have different parameters, such as, for example, different slopes, different offsets, etc. In various examples, the approximations illustrated by lines 4002, 4004, 4006, 4008 may be used by the analyte sensing system over the indicated range of in-vivo sensor temperatures _TWE . For example, FIG. 41 is a flow chart illustrating an example of a process flow 4100 that may be performed by an analyte sensor system to use multiple temperature models over different ranges of in-vivo sensor temperatures _TWE , as illustrated by FIG. 40, for example. In operation 4102, the analyte sensor system may determine the in-vivo sensor temperature _TWE using the current model. In operation 4104, the analyte sensor system may determine a temperature compensated sensitivity, M(t) _comp , using the in-vivo sensor temperature. One exemplary method of determining a temperature compensated sensitivity using the in-vivo sensor temperature is described herein with respect to equations [3]-[4]. In act 4106 , the analyte sensor system determines an estimated analyte value using the temperature compensated sensitivity determined in act 4104 .

In optional operation 4108, the analyte sensor system determines a metric derived from the in-vivo sensor temperature. Examples of such metrics include the rate of change of the in-vivo sensor temperature and/or the temperature residual as described herein. In operation 4110, the analyte sensor system determines whether the in-vivo sensor temperature _TWE and/or one of the metrics determined in operation 4106 indicates that a new model should be used to determine the in-vivo sensor temperature. For example, if the latest in-vivo sensor temperature is in a different range than the range of the model currently being used, a new model may be indicated. Using FIG. 40 as an example, if the in-vivo sensor temperature _TWE was calculated using a model corresponding to line 4004, but is between temperature 4012 and temperature 4014, the use of the model corresponding to line 4002 may be indicated.

In some examples, if one of the metrics determined in operation 4204 is outside of a predetermined range, a new model may be indicated. This may occur, for example, if the analyte sensor system utilizes different temperature models over different ranges of in vivo sensor temperature change rates and/or temperature residuals.

If a change to the temperature model is indicated in operation 4110, the analyte sensor system may set a new model to be used for the next sample from the in vitro temperature sensor in operation 4112. If no model modification is indicated, the analyte sensor system may maintain the previously used temperature model for the next sample from the in vitro temperature sensor in operation 4114.

FIG. 42 is a flow chart illustrating another example of a process flow 4200 that may be performed by an analyte sensor system to use multiple temperature models over different ranges of the in-vivo sensor temperature _TWE , for example, as illustrated by FIG. 40. In operation 4202, the analyte sensor system may determine the in-vivo sensor temperature _TWE using the current model. In optional operation 4204, the analyte sensor system determines a metric derived from the in-vivo sensor temperature. Examples of such metrics include the rate of change of the in-vivo sensor temperature and/or the temperature residual as described herein. In operation 4206, the analyte sensor system determines whether the in-vivo sensor temperature _TWE and/or one of the metrics determined in operation 4206 indicates that a new model should be used to determine the in-vivo sensor temperature. For example, if the latest in-vivo sensor temperature and/or metric derived in operation 4204 is in a different range than the range of the model currently being used, as described herein, a new model may be indicated.

If a change to the temperature model is indicated in operation 4206, the analyte sensor system may re-determine the in-vivo sensor temperature _TWE using the updated model. If no model change is indicated and/or after determining the new in-vivo sensor temperature _TWE using the updated model, the analyte sensor system may

In act 4210, the analyte sensor system may determine a temperature compensated sensitivity, M(t) _comp , using the in-vivo sensor temperature determined in act 4202 or act 4204. In act 4212, the analyte sensor system determines an estimated analyte value using the temperature compensated sensitivity determined in act 4210.

As described herein, the in vivo sensor temperature may be used to generate a temperature-compensated sensitivity, M(t) _comp , that allows the analyte sensor system to take into account the effects of temperature when determining an estimated analyte value. In some instances, such as when the analyte is glucose, factors other than sensitivity may be affected by the in vivo sensor temperature. Equation [5] above showed one way to determine an estimated glucose value using temperature-compensated sensitivity. Equation [29] shows a higher order approximation of the relationship shown in equation [5] that includes additional temperature-dependent factors.

In equation [29], M(t) is a sensitivity that may be temperature compensated as described herein. In this example, _B2 (t) is a parametric model that predicts the in vivo glucose bias or the difference between the host's blood glucose and interstitial fluid glucose concentrations. For example, _B2 (t) may correspond to the offset in equation [5]. _B1 (t) is a parametric model that predicts the non-glucose background signal. The non-glucose background signal may be caused, for example, by non-glucose substances diffusing through an interference layer of the analyte sensor. However, the diffusion rate may be temperature dependent. Thus, in some examples, the accuracy of the estimated glucose value may be increased by generating a temperature-compensated non-glucose signal _B1 (t).

The temperature compensated non-glucose signal B ₁ (t) may be generated using the in-vivo sensor temperature T _WE determined as described herein. In some examples, the temperature compensated non-glucose signal B ₁ (t) may be given by the following equation [30]:

In equation [30], T _ref is the reference temperature. The value d is the percent change in non-glucose signal per degree Celsius. The value of d may be determined from bench testing for a particular sensor configuration. In some examples, d may be modeled as a function of time, d(t), where t may be measured from the start of the sensor session.

43 is a flow chart illustrating an example of a process flow 4300 that may be performed by an analyte sensor system, such as a glucose sensor system, to determine a temperature compensated estimated glucose value. In the example process flow 4300, a signal from an in vitro temperature sensor is upsampled in that it is used to generate two or more temperature compensated analyte values based on two or more sampled values from the analyte sensor. In some examples, the sampling period of the in vitro temperature sensor may be twice the period of the analyte sensor such that each in vivo sensor temperature _TWE is used in conjunction with two samples of the analyte sensor. For example, the sampling period of the in vitro temperature sensor may be 60 seconds while the sampling period of the analyte sensor may be 30 seconds. Other values and ratios may also be used.

Referring to process flow 4300, in operation 4302, the glucose sensor system determines a current in-vivo sensor temperature _TWE , for example, using the measured in-vitro temperature _TWE . The in-vivo sensor temperature _TWE may be determined using, for example, any of the temperature models and associated techniques described herein.

In operation 4304, the glucose sensor system determines a temperature compensated sensitivity using the in-vivo sensor temperature _TWE determined in operation 4302. The temperature compensated sensitivity may be determined in any suitable manner, for example, as described herein with respect to equations [2]-[5]. In operation 4306, the glucose sensor system determines a temperature compensated non-glucose signal using the in-vivo sensor temperature. The temperature compensated non-glucose signal may be determined, for example, as described herein with respect to equation [30]. In operation 4308, the glucose sensor system determines an estimated glucose value using the temperature compensated sensitivity and the temperature compensated non-glucose signal. For example, the glucose sensor system may determine an estimated glucose value using equation [29] above.

In some examples, it may be desirable to sample an analyte sensor of an analyte sensor system at a different sampling rate than an in vitro temperature sensor, such as one of the temperature sensors 240, 242, 268, 270 described herein. For example, in some examples, a temperature model for converting in vitro temperature to in vivo sensor temperature may be optimized to run at a first period T1, while an analyte model may be optimized to run at a different period T2. Thus, in some examples, the analyte sensor system may be configured to upsample the sensor signal from the analyte sensor and/or the in vitro signal from the in vitro temperature sensor to match the periods of the two sensors.

44 is a flow chart illustrating an example of a process flow 4400 that may be performed by an analyte sensor system to upsample a signal from an in vitro temperature sensor. In operation 4402, the analyte sensor system accesses a first sample from the in vitro temperature sensor. In operation 4404, the analyte sensor system determines the in vivo sensor temperature _TWE using the sample from the in vitro temperature sensor. The in vivo sensor temperature _TWE may be determined using any of the example techniques described herein.

In operation 4406, the analyte sensor system accesses the first analyte sensor value. In operation 4408, the analyte sensor system determines a first estimated analyte value using the in vivo sensor temperature T _WE determined in operation 4404 and the first analyte sensor value 757 accessed in operation 4406. For example, the analyte sensor system may generate the first estimated analyte value using a temperature compensated sensitivity and/or a temperature compensated non-glucose signal as described herein.

In operation 4408, the analyte sensor system accesses a second analyte sensor value captured after the first analyte sensor value. In operation 4412, the analyte sensor system determines a second estimated analyte value using the first analyte sensor value accessed in operation 4410 and the in-vivo sensor temperature _TWE determined in operation 4404. For example, the analyte sensor system may generate the first estimated analyte value using a temperature compensated sensitivity and/or a temperature compensated non-glucose signal as described herein. In some examples, the temperature compensated sensitivity and/or the temperature compensated non-glucose signal generated for operation 4408 may be used again in operation 4412. In other examples, the analyte sensor system may generate a different temperature compensated sensitivity and/or non-glucose signal in operation 4412 despite using the in-vivo sensor temperature _TWE determined in operation 4404.

FIG. 45 is a schematic diagram of another exemplary analyte sensor system 4500 engaged with tissue of a host. The analyte sensor system 4500 includes a housing 4506 and an adhesive pad 4504 for securing the housing 4506 (e.g., a base thereof) to the skin 4508 of the host. FIG. 46 is a cross-sectional view of the exemplary configuration of FIG. 45 along line AA. FIG. 46 shows a circuit board 4512 within the housing 4506. The circuit board 4505 includes electronic components that may be part of the sensor electronics described herein. The sensor 4524 is electrically coupled to the circuit board 4512 and extends outside the housing 4506 and is inserted into the skin 4508 of the host. The sensor 4524 may also be mechanically coupled to the housing 4506, the circuit board 4512, and/or another suitable component. FIG. 46 also shows a battery 4514 for providing power to the sensor electronics, including, for example, the components on the circuit board 4512, and battery cage components 4516, 4518 for securing the battery 4514 within the housing 4506.

FIG. 46 also illustrates an exemplary temperature sensor 4520 in electrical communication with the circuit board 4512. For example, the temperature sensor 4520 can be configured to provide an indication of temperature to a component of the sensor electronics on the circuit board 4512. In some examples, the temperature sensor 4520 is also mechanically coupled to the circuit board 4512. The temperature sensor 4520 can be used similarly to the other temperature sensors 240, 242, 268, 270 described herein.

In the example of FIG. 46, a thermally conductive material 4522 is provided between the temperature sensor 4520 and the host's skin 4508. The thermally conductive material 4522 is positioned to conduct heat from the host's skin 4508 to the temperature sensor 4520. In this manner, the temperature indication generated by the temperature sensor 4520 may be more indicative of the temperature of the host's skin 4508 and less indicative of the ambient temperature. The thermally conductive material 4522 may include, for example, a metal or other suitable thermal conductor. The thermally conductive material 4522 may be cylindrical, cubic, rectangular prism shaped, or may have any other suitable shape. The thermally conductive material 4522, in some examples, extends through the adhesive pad 4504 and is in direct contact with the skin 4508. In other examples, the thermally conductive material 4522 is in contact with one or more intermediate layers between the thermally conductive material 4522 and the skin 4508, such as, for example, the adhesive pad 4504, a layer of the housing 4506, etc.

Any of the methods described or illustrated herein may include delivering a therapy, such as delivering insulin (e.g., using a wearable pump or smart pen) based at least in part on the determined temperature-compensated glucose concentration level. For example, the temperature-compensated glucose level may be provided to a pump, smart pen, or other device that may determine a therapy using the temperature-compensated glucose level. The methods may also be combined (e.g., in serial or parallel form) or mixed to form an aggregate method that combines two or more methods.

The systems, devices, and methods described herein may be applied to any type of analyte sensor or any type of glucose sensor. Any specific reference to a "glucose sensor" or "analyte sensor" or "glucose monitor" should be understood to be applicable to any glucose sensor, analyte sensor, glucose monitor, or other sensor that is affected by temperature. For example, methods described in the context of a glucose sensor are also applicable to other types of analyte sensors.

Although the methods for evaluating or correcting temperature measurements have been described in the context of physiological sensors and temperature compensation, the methods may also be applied in other contexts where temperature information and the accuracy of temperature information are relevant. For example, the methods may be applied to the use of temperature devices in smart devices such as handheld devices, smartphones, vehicles, watches, smart glasses, or other wearable devices.

Each of these non-limiting examples may stand on its own or may be combined in various permutations or combinations with one or more of the other examples.

The above detailed description includes reference to the accompanying drawings, which form a part of the detailed description. The drawings show, by way of illustration, specific embodiments in which the invention may be practiced. These embodiments are also referred to herein as "examples." Such examples may include elements in addition to those shown or described. However, the inventors also contemplate examples in which only those elements shown or described are provided. Moreover, the inventors also contemplate examples that use any combination or permutation of those elements (or one or more aspects thereof) shown or described with respect to a particular example (or one or more aspects thereof), or with respect to other examples (or one or more aspects thereof) shown or described herein.

In the event of a conflict of usage between this document and any document incorporated by reference, the usage in this document takes precedence.

In this document, the terms "a" or "an" are used to include one or more, as is common in patent documents, regardless of other instances or usages of "at least one" or "one or more". In this document, the term "or" is used to refer to a non-exclusive, such that "A or B" includes "A but not B", "B but not A", "A and B", unless otherwise indicated. In this document, the terms "including" and "in which" are used as the plain English equivalents of the respective terms "comprising" and "wherein". Also, in the following claims, the terms "including" and "comprising" are open-ended, i.e., a system, device, article, composition, formulation, or process that includes elements in addition to those listed after such terms in a claim is still considered to be within the scope of that claim. Moreover, in the following claims, the terms "first," "second," and "third," etc., are used merely as labels and are not intended to impose numerical requirements on their objects.

Geometric terms such as "parallel," "perpendicular," "circular," "square," and the like are not intended to require absolute mathematical precision unless the context indicates otherwise. Instead, such geometric terms allow for variations due to manufacturing or equivalent functions. For example, if an element is described as "circular" or "approximately circular," components that are not exactly round (e.g., those that are slightly elliptical or that are multi-sided polygons) are still encompassed by this description.

The method examples described herein may be at least partially implemented on a machine or computer. Some examples may include a computer-readable medium or machine-readable medium encoded with instructions operable to configure an electronic device to perform the methods described in the examples above. Implementations of such methods may include code, such as microcode, assembly language code, high-level language code, etc. Such code may include computer-readable instructions for performing various methods. This code may form part of a computer program product. Further, in examples, the code may be tangibly stored on one or more volatile, non-transitory, or non-volatile tangible computer-readable media, such as during execution or at other times. Examples of these tangible computer-readable media may include, but are not limited to, hard disks, removable magnetic disks, removable optical disks (e.g., compact disks and digital video disks), magnetic cassettes, memory cards or sticks, random access memory (RAM), read only memory (ROM), etc.

The above description is intended to be illustrative, not restrictive. For example, the above examples (or one or more aspects thereof) may be used in combination with each other. Other embodiments may be used, such as those by one of ordinary skill in the art upon reviewing the above description. The Abstract is provided to comply with 37 CFR § 1.72(b) to enable the reader to quickly grasp the nature of the technical disclosure. It is submitted with the understanding that it will not be used to interpret or limit the scope or spirit of the claims. Also, in the above "Detailed Description of the Invention," various features may be grouped together to streamline the disclosure. This should not be construed as intending that any unclaimed disclosed feature is essential to any claim. Rather, the subject matter of the invention may lie in less than all features of a particular disclosed embodiment. Thus, the following claims are incorporated into the "Detailed Description of the Invention" as examples or embodiments, and each claim stands on its own as a separate embodiment, and it is contemplated that such embodiments may be combined with each other in various combinations or permutations. The scope of the invention should be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled.

Claims (28)

1. An analyte sensor system for generating an estimated analyte value, comprising:
an in vitro temperature sensor;
an in vivo analyte sensor;
and at least one processor programmed to perform operations, the operations comprising:
accessing a first sensor signal from the in vivo analyte sensor;
accessing a first temperature signal from the ex-vivo temperature sensor;
determining that the first temperature signal satisfies a temperature signal condition;
generating a first analyte sensor temperature based at least in part on the first temperature signal after determining that the first temperature signal satisfies a temperature signal condition;
determining that the first analyte sensor temperature at least partially satisfies a temperature condition by determining that the first analyte sensor temperature is within a first temperature range;
generating a first temperature compensated sensitivity based at least in part on the first analyte sensor temperature;
generating a first estimated analyte value based at least in part on the first sensor signal and the first temperature compensated sensitivity. Determining that the first analyte sensor temperature satisfies the temperature condition;
generating a temperature rate of change using the first analyte sensor temperature and at least one previous analyte sensor temperature indicated by a previous temperature signal;
and determining that the temperature rate of change satisfies a rate of change condition. The operation,
generating a first ex-vivo temperature using the first temperature signal;
The analyte sensor system of claim 1 , further comprising: generating the first analyte sensor temperature using the first in-vitro temperature. The analyte sensor system of claim 3, wherein determining that the first in vitro temperature indicated by the first temperature signal satisfies the temperature condition includes determining that a difference between the first in vitro temperature and the first analyte sensor temperature is less than a threshold value. The operation,
determining that a second analyte sensor temperature as indicated by a second temperature signal from the in vitro temperature sensor does not satisfy the temperature condition;
10. The analyte sensor system of claim 1, further comprising: in response to determining that the second analyte sensor temperature indicated by the second temperature signal does not meet the temperature condition, performing a response action. The analyte sensor system of claim 1, wherein the operations further include determining that the first temperature signal satisfies a temperature signal condition before generating the first temperature compensated sensitivity. 1. A glucose sensor system for generating an estimated glucose value, comprising:
an in vitro temperature sensor;
an in vivo analyte sensor;
and at least one processor programmed to perform operations, the operations comprising:
Accessing a first sensor signal from an in-vivo glucose sensor;
accessing a first temperature signal from the ex-vivo temperature sensor;
generating a first glucose sensor temperature based at least in part on the first temperature signal;
generating a first temperature compensated sensitivity based at least in part on the first glucose sensor temperature;
generating a first temperature compensated non-glucose signal based at least in part on the first glucose sensor temperature;
generating a first estimated glucose value based at least in part on the first sensor signal, the first temperature compensated sensitivity, and the first temperature compensated non-glucose signal. The operation,
Accessing a second sensor signal from the in-vivo glucose sensor;
accessing a second temperature signal from the ex-vivo temperature sensor;
generating a second glucose sensor temperature based at least in part on the second temperature signal;
determining that the second glucose sensor temperature does not satisfy a temperature condition;
generating a second temperature compensated non-glucose signal based at least in part on the default temperature;
8. The glucose sensor system of claim 7, further comprising: generating a second estimated glucose value based at least in part on the second sensor signal and the second temperature compensated non-glucose signal. Producing the first temperature compensated non-glucose signal comprises:
determining a difference between the first glucose sensor temperature and a reference glucose sensor temperature;
and applying a non-glucose compensation factor to the difference. 8. The glucose sensor system of claim 7, wherein the operations further include determining that the first glucose sensor temperature satisfies a temperature condition prior to generating the first temperature compensated sensitivity, and determining that the first glucose sensor temperature satisfies the temperature condition includes determining that the first glucose sensor temperature is within a first temperature range. The operation,
Prior to generating the first temperature compensated sensitivity, further comprising determining that the first glucose sensor temperature satisfies a temperature condition, wherein determining that the first glucose sensor temperature satisfies the temperature condition comprises:
generating a temperature change rate using the first glucose sensor temperature and at least one previous glucose sensor temperature indicated by a previous temperature signal;
and determining that the temperature rate of change does not exceed a rate of change condition. The operation,
Prior to generating the first temperature compensated sensitivity,
determining that the first glucose sensor temperature satisfies a temperature condition;
generating a first ex-vivo temperature using the first temperature signal;
The glucose sensor system of claim 7 , further comprising: using the first in vitro temperature to generate the first glucose sensor temperature. The operation,
accessing a second sensor signal from the in vivo glucose sensor;
accessing a second temperature signal from the ex-vivo temperature sensor;
generating a second glucose sensor temperature based at least in part on the second temperature signal;
determining that the second glucose sensor temperature does not satisfy a temperature condition;
8. The glucose sensor system of claim 7, further comprising: in response to determining that the second glucose sensor temperature exceeds the temperature condition, performing a response action. The response action is:
generating a second temperature compensated sensitivity based at least in part on the default glucose sensor temperature;
and generating a second estimated glucose value based at least in part on the second sensor signal and the second temperature compensated sensitivity. The response action is:
pausing the display of the glucose concentration on a display associated with the glucose sensor system;
14. The glucose sensor system of claim 13, comprising at least one of: generating a second estimated glucose value based at least in part on the second sensor signal and a temperature uncompensated sensitivity; or determining that the first temperature signal satisfies a temperature signal condition. 1. A method for generating an estimated glucose concentration, comprising:
Accessing a first sensor signal from an in-vivo glucose sensor;
Accessing a first temperature signal from an ex-vivo temperature sensor;
generating a first glucose sensor temperature based at least in part on the first temperature signal;
generating a first temperature compensated sensitivity based at least in part on the first glucose sensor temperature;
generating a first temperature compensated non-glucose signal based at least in part on the first glucose sensor temperature;
generating a first estimated glucose concentration based at least in part on the first sensor signal, the first temperature compensated sensitivity, and the first temperature compensated non-glucose signal. Accessing a second sensor signal from the in-vivo glucose sensor;
accessing a second temperature signal from the ex-vivo temperature sensor;
generating a second glucose sensor temperature based at least in part on the second temperature signal;
determining that the second glucose sensor temperature does not satisfy a temperature condition;
generating a second temperature compensated non-glucose signal based at least in part on the default temperature;
generating a second estimated glucose concentration based at least in part on the second sensor signal and the second temperature compensated non-glucose signal;
The method of claim 16 further comprising: Producing the first temperature compensated non-glucose signal comprises:
determining a difference between the first glucose sensor temperature and a reference glucose sensor temperature;
and applying a non-glucose compensation factor to the difference. 17. The method of claim 16, further comprising determining that the first glucose sensor temperature meets a temperature condition prior to generating the first temperature compensated sensitivity. 20. The method of claim 19, wherein determining that the first glucose sensor temperature satisfies the temperature condition includes determining that the first glucose sensor temperature is within a first temperature range. determining that the first glucose sensor temperature satisfies the temperature condition;
generating a temperature change rate using the first glucose sensor temperature and at least one previous glucose sensor temperature indicated by a previous temperature signal;
determining that the temperature rate of change does not exceed a rate of change condition. accessing a second sensor signal from the in vivo glucose sensor;
accessing a second temperature signal from the ex-vivo temperature sensor;
generating a second glucose sensor temperature based at least in part on the second temperature signal;
determining that the second glucose sensor temperature does not satisfy a temperature condition;
performing a response action in response to determining that the second glucose sensor temperature exceeds the temperature condition; and
The method of claim 16 further comprising: The response action is:
generating a second temperature compensated sensitivity based at least in part on the default glucose sensor temperature;
and generating a second estimated glucose concentration based at least in part on the second sensor signal and the second temperature compensated sensitivity. 23. The method of claim 22, wherein the response action includes pausing the display of the glucose concentration on a display associated with the glucose sensor. 23. The method of claim 22, wherein the response action includes generating a second estimated glucose concentration based at least in part on the second sensor signal and a temperature uncompensated sensitivity. receiving a second sensor signal from the in-vivo glucose sensor;
generating a second temperature compensated sensitivity based at least in part on the first glucose sensor temperature;
generating a second estimated glucose concentration based at least in part on the second sensor signal and the second temperature compensated sensitivity;
The method of claim 16 further comprising: 17. The method of claim 16, further comprising determining that the first temperature signal satisfies a temperature signal condition prior to generating the first temperature compensated sensitivity. 28. The method of claim 27, wherein the determining that the first temperature signal satisfies a temperature signal condition is performed by the in vitro temperature sensor.