CN115736913A - Noninvasive blood glucose detection method and system - Google Patents

Abstract

The invention relates to the technical field of blood sugar detection, in particular to a noninvasive blood sugar detection method and system. The fingertip of a person to be tested is inserted into the detection probe, then the light source generator is controlled to alternately emit light sources with various different wavelengths, light penetrating through the fingertip is collected and converted into an electric signal, then the blood oxygen saturation, the heart rate, the body surface temperature of the fingertip, the heat radiation energy of the fingertip and the environment temperature of the fingertip of the person to be tested are collected, the information is used as input data to be transmitted into the upper computer after data processing, and the upper computer obtains the blood sugar value according to a trained model. The model of the upper computer needs to collect the invasive blood sugar value of a person to be detected to carry out model calibration, and the model after the calibration is qualified can carry out noninvasive blood sugar detection on the patient.

Description

Method and system for non-invasive blood glucose detection

technical field

The invention relates to the technical field of blood sugar detection, in particular to a non-invasive blood sugar detection method and system.

Background technique

Diabetes is a complex chronic disease, 50%-70% of patients have no obvious symptoms during the disease, but the chronic complications it causes develop unknowingly, so people call it the "invisible killer", especially With the improvement of people's living standards, the number of diabetic patients has increased dramatically.

The blood sugar level of diabetic patients will fluctuate greatly within a day. In order to detect the blood sugar level during treatment, standard blood sugar testing needs to be carried out at least seven times a day. In daily life, if diabetics can regularly detect blood sugar concentration, and cooperate with drugs at the same time, they can control their blood sugar concentration well. Therefore, it is very important for diabetics to detect blood sugar levels at any time. However, the traditional blood glucose test is to measure blood glucose through venous blood and capillary blood. This method is time-consuming and costly. Although there are some portable blood glucose detectors, the instrument still requires fingertip blood collection, which is inconvenient to use.

In recent years, the rise of non-invasive blood sugar detection technology has made non-invasive blood sugar detection methods gradually known to people. Most of the non-invasive blood sugar detection methods derive blood sugar values by measuring the reflection, diffuse reflection and transmission of light when light passes through the human body. However, non-invasive blood glucose There has always been a problem of measurement accuracy in detection. After light penetrates the human body, other blood tissues in the human body will also absorb light, which results in weak signals and strong interference during non-invasive blood glucose detection. In particular, the skin and blood oxygen saturation of people of different age groups are very different, which further leads to inaccurate measurement results.

The existing non-invasive blood sugar method measures the transmittance of different light through the human body, and then obtains the heart rate of the detected person; obtains the body temperature of the detected person; obtains the body radiation heat of the detected person; and according to the blood sugar Index data, heart rate, body temperature, body radiant heat, and calculate the blood sugar level of the person being tested. However, how to reduce factor interference and improve individual measurement accuracy remains to be studied.

Contents of the invention

The purpose of the present invention is to provide a non-invasive blood glucose detection method and system to solve the above-mentioned problems.

The purpose of the present invention can be achieved through the following technical solutions:

A noninvasive blood glucose detection method, said blood glucose detection method comprising the following steps:

Step 1: Insert the fingertip of the subject into the detection probe, and control the light source generator to alternately emit light sources of two different wavelengths;

Step 2: The optical signal acquisition module of the hardware system collects the transmitted light that penetrates the fingertip, and uses the data acquisition module to collect body surface temperature, ambient temperature, blood oxygen saturation and body surface radiation energy values;

Step 3: Perform signal processing on the collected data to remove interference;

Step 4: The processed signal is transmitted to the host computer, and the host computer calculates the blood glucose level according to the preset program model;

Step 5: Visually display the blood glucose value.

As a further description of the solution of the present invention, in the step 1, two light sources of different wavelengths are selected as red light of 400nm-700nm and infrared light of 700nm-2500nm, and the blood sugar level is calculated according to the two light sources of different wavelengths.

As a further description of the solution of the present invention, the working steps of the hardware system in the step 2 are as follows:

Step 21: Put the finger of the person under test into the blood oxygen finger clip device, and hold it stably;

Step 22: The embedded microprocessor controls the light source generator to alternately emit 400nm-700nm red light and 700nm-2500nm infrared light;

Step 23: The optical signal acquisition module collects the light that penetrates the fingertip, converts the optical signal into an electrical signal, performs signal processing on the electrical signal, and transmits the processed signal to the host computer;

Step 24: The parameter information collection module collects blood oxygen saturation, heart rate, body surface temperature of fingertips, thermal radiation energy of fingertips, and ambient temperature and transmits the signals to the host computer.

As a further description of the solution of the present invention, the hardware system circuit needs both an analog power supply and a digital power supply, so an isolation circuit is designed around the power supply to ensure the stability of the power supply module.

As a further description of the solution of the present invention, the steps for calibrating the model in step 4 are as follows:

Step 41: Obtain parameters such as heart rate and body surface temperature based on the blood glucose standard value measured invasively;

Step 42: Diversify parameters and blood glucose standard values;

Step 43: The training set is used to train and calibrate the model to obtain a calibrated model;

Step 44: Verify the calibrated model with a verification set;

Step 45: If the verification result meets the standard, non-invasive blood glucose testing can be performed; if the standard is not met, repeat steps 41-44.

As a further description of the solution of the present invention, in the step 44, the correlation coefficient, the prediction standard deviation and the correction standard deviation are used to evaluate the quality of the model.

A non-invasive blood sugar detection system, the system includes: an embedded microprocessor, a light source generator, an optical signal acquisition module, a parameter information acquisition module and a communication module.

As a further description of the solution of the present invention, the embedded microprocessor: together with the light source generator, optical signal acquisition module, parameter information acquisition module and communication module, is used to control the normal operation of the entire system;

Light source generator: connected with the embedded microprocessor, it is used to alternately emit light sources of two different wavelengths;

Optical signal acquisition module: connected with the embedded microprocessor, used to collect the transmitted light penetrating the fingertip;

Parameter information collection module: connected with the embedded microprocessor, used to collect body surface temperature, ambient temperature, blood oxygen saturation and body surface radiation energy value;

Communication module: connected with the embedded microprocessor, used to send the collected data to the host computer for analysis.

Beneficial effects: 1. The present invention alternately irradiates various light sources of 400nm-700nm red light and 700nm-2500nm infrared light, adopts transmission or reflection spectroscopy, and embedded microprocessor drives red light and infrared light emission respectively The end irradiates the fingertips of the human body, and the photoelectric receiving end is responsible for collecting the transmitted light, and then the signal is amplified and filtered and photoelectrically converted through the information processing module, and the transmission ratio of two different wavelengths of light and the initial blood sugar level are calculated by the formula.

2. This method also uses energy metabolism conservation method to correct the blood glucose measurement value, obtains the correction parameters through the hardware system, and uses the correction parameters to correct the blood glucose value to obtain a more accurate blood glucose value. The correction parameters include human blood oxygen saturation, human body Heart rate value, fingertip body surface temperature value, ambient temperature value and body radiation energy value, the correction method adopts artificial neural network, after inputting the data of the subject to be tested, the model can be corrected to obtain a blood sugar detection model suitable for individuals, and then Improve the accuracy of blood sugar detection.

3. The non-invasive blood glucose detection system designed in this paper integrates technologies such as photoelectricity, chemistry, medicine and artificial intelligence, and has strong reliability and a strong theoretical basis. Moreover, the system can satisfy the majority of diabetic patients to perform non-invasive, real-time, continuous high-precision self-testing of blood sugar concentration at home, and the cost is lower and the volume is smaller.

Description of drawings

The present invention will be further described below in conjunction with the accompanying drawings.

Fig. 1 is a schematic flow diagram of a part of a non-invasive blood sugar detection method provided by the present invention;

Fig. 2 is the working schematic diagram of hardware system of the present invention;

Fig. 3 is a calibration flowchart of the host computer model of the present invention.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

Please refer to Fig. 1, the present invention is a non-invasive blood glucose detection method, characterized in that the blood glucose detection method comprises the following steps:

Step 1: Insert the fingertip of the subject into the detection probe, and control the light source generator to alternately emit light sources of two different wavelengths;

Step 2: The optical signal acquisition module of the hardware system collects the transmitted light that penetrates the fingertip, and uses the data acquisition module to collect body surface temperature, ambient temperature, blood oxygen saturation and body surface radiation energy values;

Step 3: Perform signal processing on the collected data to remove interference;

Step 4: The processed signal is transmitted to the host computer, and the host computer calculates the blood glucose level according to the preset program model;

Step 5: Visually display the blood glucose value.

The key parameters are transmitted light, body surface temperature, ambient temperature, blood oxygen saturation and body radiant energy values are correction parameters for calculating blood sugar. The flowchart of this method is shown in Fig. 1 .

In the step 1, the formula for obtaining blood glucose concentration through transmitted light is as follows:

Where ε ₀ is the total absorption rate in the tissue, c ₀ is the concentration of light-absorbing substances, and L is the length of the light path; ε ₁ is the absorption coefficient of blood sugar in arterial blood, and c ₁ is the concentration of light-absorbing substances in blood sugar in arterial blood; ε ₂ is the absorption coefficient of the reference substance in the arterial blood, and c ₂ is the light-absorbing substance concentration of the reference substance in the arterial blood.

Since the blood volume in the blood vessel will change when the heart beats, which will cause the change of the optical path length, ΔL is used to represent the variable of the optical path length. Introducing ΔL into formula 1 will cause the light intensity I to become I+ΔI. Since ΔI is much smaller than I, it can be obtained after transformation:

Since the single-wavelength spectrum is prone to interference and cause deviation when measuring blood glucose, this method adopts dual wavelengths for non-invasive detection of blood glucose, the wavelengths are λ ₁ and λ ₂ respectively, and the formula 2 is expressed as

Since the selected reference substance has the same absorption coefficient at two different wavelengths, it is known that

Subtract formulas 3 and 4, and transform the formula to get

记为R，则

Will

denoted as R, then

The λ ₁ that this method selects is the red light of 400nm-700nm as reference wavelength, and λ ₂ is 700nm-2500nm infrared light as blood sugar sensitive wavelength, HbO for reference material guarantees that the absorption coefficient of two different wavelengths is identical.

In this method, the energy metabolism conservation method is used to correct the non-invasive blood glucose value, and the energy metabolism conservation method is used to correct the non-invasive blood glucose value. Thermal radiation and ambient temperature.

The fingertip is selected as the measurement site of this method, and the fingertip is richer in capillaries than other parts, and the individual fingertips of different ages and genders have little difference.

Please refer to Fig. 2, a schematic diagram of the hardware system required for completing the method, a non-invasive blood glucose detection system, characterized in that the system includes: an embedded microprocessor, a light source generator, and an optical signal acquisition module , parameter information collection module and communication module.

The embedded microprocessor: together with the light source generator, optical signal acquisition module, parameter information acquisition module and communication module, is used to control the normal operation of the entire system;

Light source generator: connected with the embedded microprocessor, it is used to alternately emit light sources of two different wavelengths;

Optical signal acquisition module: connected with the embedded microprocessor, used to collect the transmitted light penetrating the fingertip;

Parameter information collection module: connected with the embedded microprocessor, used to collect body surface temperature, ambient temperature, blood oxygen saturation and body surface radiation energy value;

Communication module: connected with the embedded microprocessor, used to send the collected data to the host computer for analysis.

The hardware system of this method needs to complete signal acquisition, signal amplification filtering, analog-to-digital conversion and communication functions, and the hardware system working steps are as follows:

Step 21: Put the finger of the person under test into the blood oxygen finger clip device, and hold it stably;

Step 22: The embedded microprocessor controls the light source generator to alternately emit 400nm-700nm red light and 700nm-2500nm infrared light;

Step 23: The optical signal acquisition module collects the light that penetrates the fingertip, converts the optical signal into an electrical signal, performs signal processing on the electrical signal, and transmits the processed signal to the host computer;

Step 24: The parameter information collection module collects blood oxygen saturation, heart rate, body surface temperature of fingertips, thermal radiation energy of fingertips, and ambient temperature and transmits the signals to the host computer.

Because the circuits in the hardware system need both digital power and analog power in the power module, the stability of the power module is very important. In this hardware system, an isolation circuit needs to be designed to ensure the stability of the power module.

Referring to Figure 3, the non-invasive blood glucose detection method of this method needs to input characteristic information, train the model in the host computer to calibrate the model, so that the model automatically outputs accurate and reliable prediction results, and the steps for calibrating the model in step 4 are as follows :

Step 41: Obtain parameters such as heart rate and body surface temperature based on the blood glucose standard value measured invasively;

Step 42: Diversify parameters and blood glucose standard values;

Step 43: The training set is used to train and calibrate the model to obtain a calibrated model;

Step 44: Verify the calibrated model with a verification set;

Step 45: If the verification result meets the standard, non-invasive blood glucose testing can be performed; if the standard is not met, repeat steps 41-44.

In step 44, the correlation coefficient (R ² ), predicted standard deviation (RMSEP) and corrected standard deviation (RMSEC) are used to evaluate the quality of the model.

The present invention can collect two wavelengths of transmitted light and parameter information through the hardware system, and transmit the collected information to the host computer, and the preset model in the host computer can predict the blood sugar of the subject according to the input parameters, realizing non-invasive blood sugar detection. Among them, the preset model of the upper computer adopts the operation method of black box. The user does not need to understand the internal model structure, and only needs to input the data according to the requirements to complete the training automatically. When the model training meets the quality evaluation standards, it can be used in practice.

An embodiment of the present invention has been described in detail above, but the content described is only a preferred embodiment of the present invention, and cannot be considered as limiting the implementation scope of the present invention. All equivalent changes and improvements made according to the application scope of the present invention shall still belong to the scope covered by the patent of the present invention.

Claims (8)

1. A non-invasive blood glucose detection method is characterized by comprising the following steps:

step 1: inserting the fingertip of a person to be detected into the detection probe, and controlling the light source generator to alternately emit a plurality of light sources with different wavelengths;

and 2, step: a light signal acquisition module of the hardware system acquires transmitted light penetrating through fingertips, and a data acquisition module is used for acquiring body surface temperature, environment temperature, blood oxygen saturation and body surface radiant energy values;

and 3, step 3: carrying out signal processing on the acquired data to remove interference;

and 4, step 4: the processed signals are transmitted into an upper computer, and the upper computer calculates blood sugar values according to a preset program model;

and 5: and visually displaying the blood sugar value.

2. A method for non-invasive blood glucose measurement according to claim 1, wherein the light sources with different wavelengths in step 1 are selected from the group consisting of red light with a wavelength of 400nm-700nm and infrared light with a wavelength of 700nm-2500nm, and the blood glucose level is calculated according to the two light sources with different wavelengths.

3. A method for non-invasive blood glucose measurement according to claim 1, wherein the hardware system in step 2 is operated as follows:

step 21: placing the finger of the measured person into a blood oxygen finger clip instrument, and stably clamping;

step 22: the embedded microprocessor controls the light source generator to alternately emit 400nm-700nm red light and 700nm-2500nm infrared light;

step 23: the optical signal acquisition module acquires light penetrating through a fingertip, converts an optical signal into an electric signal, performs signal processing on the electric signal, and transmits the processed signal to the upper computer;

step 24: the parameter information acquisition module collects the blood oxygen saturation, the heart rate, the body surface temperature of the fingertip, the heat radiation energy of the fingertip and the environment temperature and transmits signals to the upper computer.

4. A method as claimed in claim 3, wherein the hardware system circuit requires both analog power supply and digital power supply, so that the power module is stable by designing an isolation circuit around the power supply.

5. The method for non-invasive blood glucose measurement according to claim 1, wherein the step of calibrating the model in step 4 comprises:

step 41: obtaining parameters such as heart rate, body surface temperature and the like according to the blood glucose standard value measured invasively;

step 42: diversity of the parameters and the blood glucose standard values is carried out;

step 43: training and calibrating the model by the training set to obtain a calibrated model;

step 44: verifying the calibrated model by using a verification set;

step 45: and (4) the verification result meets the standard, noninvasive blood glucose detection can be carried out, and the steps 41 to 44 are repeated when the standard is not met.

6. A non-invasive method of blood glucose measurement according to claim 5, wherein the correlation coefficient, the prediction standard deviation and the correction standard deviation are used to validate and evaluate the model in step 44.

7. A non-invasive blood glucose detection system using the method of any one of claims 1-6, the system comprising: the device comprises an embedded microprocessor, a light source generator, a light signal acquisition module, a parameter information acquisition module and a communication module.

8. The system of claim 7, wherein the embedded microprocessor: the optical signal acquisition module is connected with the light source generator, the optical signal acquisition module, the parameter information acquisition module and the communication module and is used for controlling the normal operation of the whole system;

a light source generator: the embedded microprocessor is connected with the light source and is used for alternately emitting two light sources with different wavelengths;

the optical signal acquisition module: the embedded microprocessor is connected with the embedded microprocessor and is used for collecting transmitted light penetrating through fingertips;

the parameter information acquisition module: the embedded microprocessor is connected with the embedded microprocessor and is used for collecting the body surface temperature, the environment temperature, the blood oxygen saturation and the body surface radiant energy value;

a communication module: and the embedded microprocessor is connected with the data acquisition unit and is used for sending the acquired data to an upper computer for analysis.

Images