JP7688593B2 - Determination device and program - Google Patents

Description

The present invention relates to a determination device, etc.

Conventionally, devices have been known that attach a sensor to the subject's body, detect specific vibrations from the sensor, and determine the subject's behavior. For example, devices are known that determine the subject's behavior, such as scratching (see, for example, Patent Document 1 and Patent Document 2).

JP 2015-112127 A JP 2021-137570 A

The present invention provides a determination device that can determine a specific state or movement of a subject by utilizing vibrations generated by the subject's movements.

In view of the above-mentioned problems, the determination device of this embodiment includes an acquisition unit that acquires a sensor value indicating vibration from a sensor placed between the bed device and the subject, and a control unit, and the control unit detects a specific vibration that is higher in frequency than vibration caused by breathing, has a higher amplitude than vibration caused by heartbeat, and is periodic, and determines that the section in which the specific vibration is detected is the section in which the subject performed a specific behavior.

The program of this embodiment also realizes, in a computer capable of acquiring sensor values indicating vibrations from a sensor placed between the bed device and the subject, a function of detecting, from the sensor values, specific vibrations that are higher frequency than vibrations caused by breathing, have higher amplitude than vibrations caused by heartbeats, and are periodic, and a function of determining that the section in which the specific vibrations are detected is the section in which the subject performed a specific behavior.

According to the present invention, it is possible to detect a specific state or movement of a subject by utilizing vibrations caused by the subject's movements.

FIG. 2 is a diagram for explaining the whole of the present embodiment. FIG. 2 is a diagram for explaining a configuration according to the present embodiment. FIG. 4 is a diagram illustrating an example of a parameter table in the present embodiment. FIG. 4 is a diagram illustrating an example of parameters in the present embodiment. 10A to 10C are diagrams for explaining specific vibration detection processing in the present embodiment. 10A to 10C are diagrams for explaining the operation of a specific vibration detection process in the present embodiment. 11 is a diagram for explaining an example of a scratching behavior determination process in the present embodiment. FIG. 11 is a diagram for explaining the operation of the scratching behavior determination process in the present embodiment. FIG. FIG. 11 is a diagram for explaining an operation example in the present embodiment. FIG. 11 is a diagram for explaining an operation example in the present embodiment. FIG. 11 is a diagram for explaining the effect of the present embodiment. FIG. 11 is a diagram for explaining the effect of the present embodiment. FIG. 13 is a diagram for explaining a modified example of the present embodiment.

There are various known methods for determining a specific state of a subject, and the specific states that can be determined are also diverse. Here, the state of the subject includes a state that involves some movement of the subject, and may include actions or behaviors consciously performed by the subject, as well as movements that occur regardless of the subject's will (involuntary movements).

The state of the subject also includes a state based on biometric information. For example, it may include whether the subject is asleep/awake based on the subject's heart rate, breathing, and body movements. The state of the subject may also include the subject's posture (such as the sleeping posture while sleeping, such as supine or lateral position, or posture such as sitting on the edge of the bed device or sitting long).

One example of a specific behavior of the subject that is included in this state of the subject is scratching behavior. Scratching behavior is a scratching behavior that is seen when the subject feels itchy or stressed, and scratching induces further itching and a worsening of the skin condition. Furthermore, it is known that skin diseases such as atopic dermatitis cause itching to become more severe at night when the skin temperature rises. Since subjects unconsciously scratch while sleeping, which hinders the improvement of skin symptoms, it has been pointed out that measuring scratching behavior during sleep is useful in treatment. For this reason, a determination device that determines a subject's scratching behavior is known.

However, the following methods are known for accurately determining whether a subject has scratched themselves. The first method involves, for example, using a video camera to film the subject while they are sleeping, and determining whether or not they have scratched themselves. The second method involves wearing a wristwatch-type actigraph (including wearable devices such as smart watches) to measure the amount of activity (amount of bodily movement quantified), which is an alternative indicator, and determining whether or not they have scratched themselves.

The first method, which involves filming with a video camera, requires that the judgement be made by a person or artificial intelligence, which places a heavy burden on the person making the judgement. In addition, as the subject is filmed, privacy issues must be taken into consideration. In the second method, which involves wearing a wristwatch-type actigraph, it has been reported that the amount of activity, which is an alternative indicator, increases with the amount of scratching. However, because turning over in bed and moving about are also included in the amount of activity, it is not necessarily an indicator that measures only scratching.

In addition, as described in JP 2015-112127 A and JP 2021-137570 A, there is a method in which a sensor device that detects vibrations is attached to the subject's body separately. With this method, the sensor device needs to be attached to the subject separately, which can be bothersome for the subject, or the sensor device can come off in the first place.

As devices that are not worn by the subject, there are known acquisition devices that can acquire biological information such as the subject's heart rate, heart rate, and breathing, and acquire the sleep state (whether the subject is asleep or awake) from changes in load detected by the sensor device by placing a sensor between the bed device (mattress) and the subject, or by placing a vibration sensor on the bed device or mattress. It was not anticipated that such acquisition devices would be able to specifically measure the continuous vibrations (body movements) that occur during scratching behavior and that are different from breathing and heart rate.

In view of these issues, one embodiment of a determination device capable of appropriately detecting a specific motion or state of a subject will be described below with reference to the drawings. Note that the following embodiment is an example of the application of the determination device of the present invention, and the scope of application of the present invention is not limited to the following embodiment.

[1. The entire system]
Fig. 1 is a diagram for explaining an overall overview of a system 1 to which a determination device is applied. As shown in Fig. 1, the system 1 is configured to include a detection device 3 placed between a floor part of a bed device 10 and a mattress 20, and a processing device 5 for processing values output from the detection device 3. The detection device 3 and the processing device 5 constitute a detection device (detection system) that determines the state of a subject. The detection device 3 and the processing device 5 may be configured as an integrated type.

When a subject (hereinafter, referred to as "subject P" as an example) is present on the mattress 20, the detection device 3 detects the subject P's body vibrations (vibrations emitted from the human body). The processing device 5 can then acquire the subject P's state based on the vibrations detected by the detection device 3. Details will be described later, but the processing device 5 can acquire the following states, for example, as the state of subject P:

- Whether the subject is lying on the bed apparatus 10 (mattress 20) or has left the bed apparatus 10 (mattress 20) - Heart rate, respiratory rate, and activity level as biological information of the subject - Posture of the subject. For example, lateral position, supine position, edge sitting position, long sitting position, etc. - Position of the subject. For example, whether the subject is in the center of the bed apparatus 10 or to one side - Movement and behavior of the subject. For example, scratching behavior as the behavior of the subject, and convulsions, trembling, etc. as actions that are not the subject's will. Also included are actions of the subject such as turning over and getting up.

In this embodiment, the calculated bioinformation values (e.g., respiratory rate, heart rate, activity level) can be output and displayed as the bioinformation values of the subject P. Note that, for example, the detection device 3 may be integrally formed with a storage unit, a display unit, etc. In addition, the processing device 5 may be a general-purpose device, and is not limited to an information processing device such as a computer, and may be configured as a device such as a tablet or smartphone.

The subjects may be people undergoing medical treatment or those in need of nursing care. They may also be healthy people who do not require nursing care, elderly people, children, people with disabilities, or even non-human animals.

The detection device 3 is configured in a sheet shape to have a small thickness. This allows it to be used without causing discomfort to the subject P even when placed between the bed device 10 and the mattress 20, making it possible to measure biometric information values in bed for a long period of time.

The detection device 3 may be capable of detecting the body vibrations of the subject P without being worn. For example, an actuator with a strain gauge or a load cell that measures the load placed on the legs of the bed may be used. Also, by using a built-in acceleration sensor, for example, it may be realized by a smartphone or tablet placed on the bed device 10. Also, one or more sensors may be provided.

In addition, in FIG. 1, the head side of the bed device 10 (mattress 20) is direction H, the foot side is direction F, and when the subject P is in a supine position, the left side is direction L and the right side is direction R.

2. Configuration
Next, the configuration of the system 1 will be described with reference to the drawings. The system 1 in this embodiment includes a detection device 3 and a processing device 5, and each functional unit (processing) may be realized by either one. In this embodiment, the processing device 5 functions as a determination device.

The processing device 5 includes a control unit 100, a signal acquisition unit 120, a memory unit 150, an input unit 160, and an output unit 170.

The control unit 100 controls the operation of the processing device 5 and is composed of one or more control devices. For example, it may be a control device such as a CPU (Central Processing Unit), or it may be a SoC (System on a Chip) with multiple functions.

The control unit 100 also realizes various processes by reading and executing various programs stored in the storage unit 150. By executing the programs, the control unit 100 may function as a specific vibration detection unit 102, a bioinformation calculation unit 104, and a subject state determination unit 106. Note that, although the control unit 100 operates as a whole in this embodiment, it may also be provided in each of the detection device 3 and the processing device 5.

The specific vibration detection unit 102 detects whether a specific vibration is occurring based on the sensor value output from the detection device 3. In other words, it detects that a specific vibration is present in the vibrations indicating the subject's body movements detected by the detection device 3. The specific vibration is a vibration that the processing device 5 wants to detect, and corresponds to a specific behavior/movement of the subject. For example, the processing device 5 can determine the subject's scratching behavior by detecting a specific vibration that corresponds to the subject's scratching behavior.

Here, specific vibrations are vibrations that are different from those caused by heartbeats and breathing. This is because the higher the frequency of vibrations caused by heartbeats (the higher the heart rate), the lower the amplitude and the lower the amplitude. Alternatively, high frequency and high heart rate are considered abnormal, and similarly, the higher the frequency of vibrations caused by breathing, the lower the amplitude and the lower the amplitude, and the breathing rate never exceeds 1 Hz (60 beats/min).

A specific vibration is data that can be obtained from body vibrations and is higher amplitude (larger vibration) than the vibrations that occur in response to heartbeats, and higher frequency (finer vibration) than the vibrations that occur in response to breathing.

For example, if the vibration (body movement) caused by breathing is 0.1 to 0.6 Hz, a specific vibration is a vibration with an amplitude equal to or greater than that of the vibration (body movement) caused by a heartbeat of 1 Hz or more (as an example, an amplitude equal to or greater than 1/5 of the maximum amplitude caused by breathing or an amplitude equal to or greater than 1/5 of the measurement range. For example, 820 or greater if the measurement range is data quantized at 12 bits (0 to 4095)). Note that vibrations caused by heartbeat and breathing will in principle result in a fluctuation in the sensor value with an amplitude within the detection range of the sensor, but a specific vibration refers to one that exceeds a specified arbitrary amplitude, and may result in a fluctuation in the sensor value with an amplitude that exceeds the detection range of the sensor.

The bioinformation calculation unit 104 calculates the bioinformation based on the signal acquired from the signal acquisition unit 120. For example, the heartbeat (heart rate) and respiration (respiratory rate) are calculated as the bioinformation of the subject. For example, the method of calculating the bioinformation may be any of the known methods described in JP 2017-047211 A (Name of invention: Bioinformation output device, bioinformation output method and program, Publication date: March 9, 2017).

The subject state determination unit 106 can determine the state of the subject. As the state of the subject, for example, it can determine whether the subject is performing a specific behavior. For example, when the subject state determination unit 106 detects a specific vibration from the specific vibration detection unit 102, it can determine that the subject is performing a specific behavior corresponding to the specific vibration.

Specifically, when the specific vibration detection unit 102 detects a specific vibration, the subject state determination unit 106 determines that the subject has performed a scratching motion.

The subject state determination unit 106 can also determine whether the subject is asleep or awake. Here, the subject's sleep state may be determined from vibration data acquired by the signal acquisition unit 120, or may be determined based on the biological information output from the biological information calculation unit 104. For example, the subject state determination unit 106 may determine whether the subject is asleep or awake from the changes in the subject's heart rate and respiratory rate. The method of determining the subject's sleep state may be, for example, a method disclosed in JP 2015-12948 A (Name of invention: Sleep evaluation device, sleep evaluation method, and sleep evaluation program, Publication date: January 22, 2015), JP 2017-213421 A (Name of invention: Sleep evaluation device and program, Publication date: December 7, 2017), etc. In addition, any other known method of determining the subject's sleep state may be used.

The subject state determination unit 106 may also determine the posture of the subject. For example, the subject state determination unit 106 may determine whether the subject is in a supine position or a lateral position. The subject state determination unit 106 may also determine whether the subject is sitting on the edge of the bed or sitting on the side. The subject state determination unit 106 may determine the posture from vibration data acquired by the signal acquisition unit 120, or may determine the posture from a change in load. As a method for determining the posture of the subject, for example, a method disclosed in JP 2019-98069 A (Name of invention: Posture determination device, Publication date: June 24, 2019) or the like may be used. In addition, any other known method for determining the posture of the subject may also be used.

The signal acquisition unit 120 acquires the signal (sensor value) output by the detection device 3. Here, the detection device 3 outputs, for example, vibration data indicating vibration as a sensor value. In this embodiment, the detection device 3 converts the change in air pressure due to vibration into a voltage value measured by a piezoelectric sensor (also, since the measurement range of the voltage value is limited, the voltage value measured varies greatly depending on individual differences, sleeping posture, and sleeping position. For example, in the case of this embodiment, it may be a voltage value in which the amplification of the sensor output is adjusted so that it does not fall outside the measurement range due to breathing movement), and outputs it. The detection device 3 may output the vibration data as an analog signal, may convert it into a digital signal, or may output a quantified version. If it is output as an analog signal, the signal received by the signal acquisition unit 120 or the control unit 100 may be converted into a sensor value. In this case, the signal acquisition unit 120 or the control unit 100 converts the vibration data into a digital voltage signal at a predetermined sampling interval. Here, if the magnitude of the signal is too large, the signal is attenuated and converted into a digital signal. Then, the signal acquisition unit 120 or the control unit 100 converts the signal into a digital signal, quantizes it, and stores it as a sensor value in the sensor value storage area 154. That is, the vibration data acquired by the signal acquisition unit 120 is sampled, quantized, and encoded by the signal acquisition unit 120 or the control unit 100, and stored as a sensor value in the sensor value storage area 154.

The storage unit 150 stores various data and programs for the processing device 5 to operate. The control unit 100 realizes various functions by reading and executing the programs stored in the storage unit 150. Here, the storage unit 150 is composed of a semiconductor memory (e.g., an SSD (Solid State Drive) or an SD card (registered trademark)), a magnetic disk device (e.g., an HDD (Hard Disk Drive)), etc. The storage unit 150 may be a built-in storage device or a removable external storage device. It may also be a storage area of an external server such as a cloud.

The memory unit 150 secures a subject data memory area 152 and a sensor value memory area 154, and stores a parameter table 156.

The subject data storage area 152 stores subject data. The subject data is data related to the subject, and each data is stored in chronological order. Examples of subject data that can be stored include the result of in-bed/out-of-bed determined by the control unit 100, bioinformation calculated by the control unit 100 from the sensor value, and the subject's state determined by the subject state determination unit 106. These data (information) can be stored in chronological order for each subject. As the state of the subject, the subject data storage area 152 may store the time (time zone) when the scratching behavior was performed, or may store the total time when the scratching behavior was performed over the course of one night. In addition, the strength of scratching (amplitude and frequency) may be calculated and stored. The strength of scratching refers to an index that comprehensively evaluates the magnitude of the scratching force and the speed of scratching. For example, the stronger the scratching force, the larger the amplitude of the specific vibration acquired, and the faster the scratching speed, the higher the frequency.

The sensor value storage area 154 stores the sensor values output by the signal acquisition unit 120 or the control unit 100. The sensor values may be stored, for example, in chronological order for each subject.

The parameter table 156 stores parameters such as thresholds and judgment values required for various processes. The parameters may be changed by the user or may be stored in advance.

Figure 3 shows an example of parameter table 156. Parameter table 156 stores as parameters the upper threshold Amax (e.g., "2458"), the lower threshold Amin (e.g., "1638"), the out-of-range staying time B (e.g., "45"), the in-range staying time C (e.g., "15"), and the detection determination value D (e.g., "128") of the set range A set to detect a specific vibration.

Although each parameter will be described in detail later, first, the set range A will be described. The set range A is a range of sensor values that the specific vibration detection unit 102 uses when determining whether or not a vibration is a specific vibration. When the specific vibration detection unit 102 goes outside (exceeds) the set range A or when the specific vibration detection unit 102 enters the set range A, it starts detection processing of the specific vibration.

Here, multiple ranges may be set for the setting range A. This is because the range (amplitude) in which the sensor value detected by the detection device 3 changes changes depending on the state of the subject, such as the subject's movement. In this embodiment, there are three settings, setting 1 to setting 3.

Figure 4 is a diagram for explaining the setting ranges. Figure 4 is a graph of the sensor values, with the horizontal axis representing time (number of samples) and the vertical axis representing the sensor values. When the sensor value is near the center, setting 1 is used. That is, setting range A is the range from upper threshold Amax-1 to lower threshold Amin-1. When the sensor value is output close to the lower limit that can be processed, setting 2 is used. That is, setting range A is the range from upper threshold Amax-2 to lower threshold Amin-2. When the sensor value is output close to the upper limit that can be processed, setting 3 is used. That is, setting range A is the range from upper threshold Amax-3 to lower threshold Amin-3.

Out-of-range time B is a parameter used to determine the time during which the sensor value is outside the set range. In-range time C is a parameter used to determine the time during which the sensor value is within the set range.

An example of the parameters in FIG. 3 is a parameter for determining scratching behavior. Therefore, parameters may be stored for each state of the subject to be determined. Also, parameters may be stored individually for each subject.

The input unit 160 accepts operational input from the user. For example, the user may perform various operational inputs, such as starting to acquire the user's status or adjusting the sensitivity of the detection device 3.

The output unit 170 outputs various types of information. For example, it may be configured as a liquid crystal display device, a light-emitting element such as an LED, a speaker that outputs sound or voice, an interface that outputs data to another recording medium, etc. Also, the input unit 160 and the output unit 170 may be configured as a touch panel by being integrated together.

The output unit 170 can not only output various information, but also display reports and conditions based on the subject's condition. For example, the output unit 170 can output the subject's motion status, such as the presence or absence of scratching behavior and the time during which the scratching behavior was performed. The output unit 170 can also output a sleep diary based on the sleep state and bioinformation. The sleep diary can output, for example, the subject's sleep/wake state for one day, and also the subject's in-bed/out-of-bed state. As a method for outputting the sleep diary, for example, a method disclosed in JP 2016-144627 A (Name of invention: Bioinformation output device, bioinformation output method and program, Publication date: August 12, 2016) or the like can be used. As a method for outputting the sleep diary, any known method can be used.

The output unit 170 may also output the scratching behavior by overlaying it on the sleep diary. This makes it possible to output the relationship between the depth of the subject's sleep and the scratching behavior. The output unit 170 may also output advice regarding the scratching behavior determined by the control unit 100. For example, the output unit 170 may output a message such as "Scratching behavior is often observed when falling asleep" by comparing with the subject's sleep state.

Also, FIG. 2 conceptually explains the configuration when a processing device 5 capable of communicating with the detection device 3 functions as a detection device. These configurations may be realized, for example, by a single device capable of detecting vibration. That is, the detection device 3 and the processing device 5 may be configured as an integrated device, or as shown in FIG. 1, the detection device 3 and the processing device 5 may be configured separately. For example, the processing device 5 may be realized by installing an application capable of realizing the functions of the processing device 5 on a smartphone. In this case, the acceleration sensor of the smartphone may be used instead of the detection device 3. Also, instead of the processing device 5, it may be realized by an external server capable of providing the same service.

The detection device 3 has a sensor that detects vibrations, or is configured to be connectable to an external sensor. The detection device 3 transmits a signal (vibration data) related to the vibrations detected by the sensor to the processing device 5 via a communication unit. The timing of transmitting the vibration data from the detection device 3 to the processing device 5 and the timing of the processing device 5 storing the vibration data may be in real time or at predetermined intervals.

The detection device 3 may have one sensor or multiple sensors. The detection device 3 may be realized, for example, by a sensor built into the actuator, or a load cell (built-in or external) provided on the frame of the bed apparatus or on the legs or casters of the bed apparatus.

The detection device 3 and the processing device 5 can be connected by wire or wirelessly. The detection device 3 may also store the vibration data in an external recording medium (e.g., an SD card). The processing device 4 may read and acquire the vibration data from the recording medium.

3. Specific vibration detection process
[3.1 Processing flow]
Next, the specific vibration detection process in this embodiment will be described with reference to Fig. 5. Hereinafter, a case where the control unit 100 detects a body vibration caused by scratching as the specific vibration will be described.

When executing the specific vibration detection process , the control unit 100 (specific vibration detection unit 102) acquires a sensor value from the detection device 3 at predetermined time intervals. That is, the control unit 100 continuously acquires the sensor value interrupted at predetermined time intervals. Note that the interval at which the control unit 100 acquires the sensor value may be synchronized with the timing of output from the detection device 3, or the control unit 100 may receive the sensor value from the detection device 3 at predetermined time intervals.

Furthermore, the control unit 100 sets each parameter when executing the specific vibration detection process . The control unit 100 reads out predetermined parameters from the parameter table 156. The parameters stored in the parameter table 156 will be described below.

The upper threshold Amax and lower threshold Amin are thresholds that determine the setting range A. The values shown in FIG. 3 are values when the vibration data is quantized to 12 bits (0 to 4095), and the data resolution, upper threshold Amax, and lower threshold Amin may be other values. Furthermore, the upper threshold Amax and lower threshold Amin may be stored as the setting range depending on the value (resolution) at which the vibration data is quantized. Furthermore, although the upper threshold Amax and lower threshold Amin are stored as values when the vibration data is quantized to 12 bits, the sensor values may be stored as is.

Out-of-range time B is a parameter that indicates the time that the sensor value is outside the set range. For example, if the time that the sensor value is above the upper threshold Amax or below the lower threshold Min exceeds out-of-range time B, it is determined that the vibration is not a specific one, and the control unit 100 can end this process.

Out-of-range staying time B indicates the judgment time, but may be expressed as the number of samples or as time. When out-of-range staying time B is the number of samples, for example, it is shown as "40" or "45" in FIG. 3. Here, the sampling interval may be, for example, 16 Hz, 64 Hz, 100 Hz, 128 Hz, 500 Hz, etc. Also, out-of-range staying time B may be shown as time, for example, "1/2" second or "1/3" second.

The in-range staying time C is a parameter that indicates the time for the waveform indicating the sensor value to transition within the set range A. For example, it indicates the time from when the sensor value, which exceeded the upper threshold Amax, falls below the upper threshold Amax to when it falls below the lower threshold Min. If the time to transition within this set range A exceeds the in-range staying time C, it is determined that it is not a specific vibration, and the control unit 100 can end this process.

The time spent within range C indicates the determination time, but may be expressed as the number of samples or as time. When the time spent within range C is the number of samples, for example, in FIG. 3, it is shown as "15" or "26". Furthermore, the time spent within range C may be shown as time, for example, as "1/5" second or "1/6" second.

The detection judgment value D is a parameter that indicates the elapsed time since it was detected that there was a possibility of a specific vibration. For example, if the detection judgment value D is exceeded from the point when the sensor value first exceeded the set range, the control unit 100 determines that the detected vibration is a specific vibration. If the control unit 100 ends this process before the detection judgment value D is reached, the vibration that was possibly a specific vibration is determined to not have been a specific vibration. The detection judgment value indicates the judgment time, but may be expressed as the number of samples or as time.

In this way, after setting the parameters read from the parameter table 156, the control unit 100 executes the following processing.

The control unit 100 initializes the sampling count to "0", the detection flag to "OFF", and the total sampling count to "0" (step S102). If the acquired sensor value exceeds the set range A, the control unit 100 executes processing from step S106 (step S104; Yes). Here, the control unit 100 determines that the set range A has been exceeded if the sensor value that was within the set range A becomes larger than the upper threshold Amax or becomes smaller than the lower threshold Amin.

The control unit 100 then executes a first determination process (steps S106 and S108). Specifically, the control unit 100 determines whether the sampling count is greater than the out-of-range stay time B (step S106). After step S106, the sampling count is incremented each time a sensor value is acquired. For example, if 10 sensor values are acquired, the sampling count becomes 10.

Then, if the number of samples exceeds the out-of-range staying time B, the control unit 100 transitions to step S118 to end the process of detecting a specific vibration (step S106; Yes → step S118).

The control unit 100 also repeats step S106 while the sensor value is outside the set range A (step S106; No → step S108; No → step S106). That is, the control unit 100 performs processing to end this process when the number of samples exceeds the out-of-range residence time B while the sensor value continues to exceed the set range A. In other words, the control unit 100 ends this process when the time during which the sensor value continues to exceed the upper threshold Amax or the time during which the sensor value continues to be below the lower threshold Amin exceeds the out-of-range residence time B.

If the sensor value is within the set range (step S108), the control unit 100 adds the sampling number to the total sampling number and initializes the sampling number to 0 (step S110). Next, the control unit 100 executes the second determination process (steps S112 and S114).

The control unit 100 determines whether the number of samples has exceeded the in-range stay time C (step S112). That is, when the sensor value is inverted (when the graph showing the sensor value transitions from a peak to a valley and from a valley to a peak), the control unit 100 determines whether the transition of the sensor value from the upper threshold Amax (or the lower threshold Amin) to the lower threshold Amin (or the upper threshold Amax) is completed within the in-range stay time C.

Here, for example, if the number of sensor value samples exceeds the range stay time C, the control unit 100 executes a process to end this process (step S112; Yes). Furthermore, the control unit 100 repeatedly executes the process until the sensor value exceeds the set range A (step S112; No → step S114; No → step S112).

If the sensor value exceeds the set range again (step S114; Yes), the control unit 100 adds the sampling number to the total sampling number and initializes the sampling number to 0 (step S116). Then, the control unit 100 executes the first determination process (steps S106 and S108). That is, the control unit 100 repeatedly executes the first process and the second process.

If the number of samples exceeds the out-of-range stay time B in step S106 (step S106; Yes) or if the number of samples exceeds the in-range stay time C (step S112; Yes), the control unit 100 transitions to step S118.

When the total number of samples is equal to or greater than the detection determination value D, the control unit 100 detects the specific vibration section (step S120). For example, the control unit 100 sets the detection flag to "1" and ends this process (step S118; Yes → step S120 → step S122). As a result, the control unit 100 determines that the current vibration is a specific vibration.

In this way, the control unit 100 detects vibrations within the range that is the start point when the sensor value exceeds the set range in step S104, and the end point is the point before the processing that is judged to be Yes in step S106 or step S1112, as a specific vibration.

In this way, according to this process, it is possible to determine the section in which a specific vibration is occurring based on the sensor value output from the sensor of the detection device 3. The section in which this specific vibration is occurring can be determined as a specific movement of the subject.

For example, in this embodiment, the specific vibration is preferably determined to be a scratching behavior of the subject. In addition, the specific vibration may be determined to be, for example, a convulsion, tremor, or shaking of the subject.

In addition, although it is preferable that the control unit 100 executes both the first process and the second process in the specific vibration detection process , it may execute only a part of the processes. For example, the control unit 100 may execute only the first process and not the second process (i.e., not compare with the in-range staying time C).

The control unit 100 may execute only the second process and not the first process (i.e., not compare with the out-of-range staying time B). The control unit 100 may also change the order of the process start in steps S104 and S108. For example, the control unit 100 may start the process when a sensor value that was greater than the upper threshold Amax becomes smaller than the upper threshold Amax, or when a sensor value that was smaller than the lower threshold Amin becomes greater than the lower threshold Amin. Alternatively, both process start patterns may be implemented at the same time. For example, if the process start order in steps S104 and S108 is changed, the control unit 100 may switch the process order in steps S106 (to S110) and S112 (to S116).

In FIG. 5, the control unit 100 determines whether the amplitude exceeds the amplitude defined in the set range A in steps S104, S108, and S114, and this may be defined as "high amplitude" in the specific vibration in this determination device. Also, the control unit 100 determines whether the data transitions within the period defined by the out-of-range staying time B and the in-range staying time C in steps S106 and S112, and if the above conditions are met, this may be regarded as a periodic vibration and defined as "high frequency" in the specific vibration in this determination device.

[3.2 Operation example]
The specific vibration detection process will be described with reference to Fig. 6. Fig. 6 is a diagram showing an example of a waveform based on a sensor value, in which the horizontal axis indicates the time axis (number of samples) and the vertical axis indicates the sensor value.

Furthermore, the upper threshold Amax and the lower threshold Amin are indicated by dashed lines as the set range A. First, the control unit 100 detects that the sensor value exceeds the upper threshold Amax at time T10. The control unit 100 detects that time T10 is the start time of the specific vibration. In other words, the control unit 100 determines that any vibration detected after this point is likely to be the specific vibration.

Then, the control unit 100 performs a first process. The control unit 100 acquires that the sensor value continues to exceed the range of the upper threshold Amax. Then, the control unit 100 acquires a time T12 at which the sensor value becomes equal to or less than the upper threshold Amax. The control unit 100 checks whether an interval R10 between times T10 and T12 is within the range of out-of-range stay time B. Then, if the interval R10 is within the range of out-of-range stay time B, the control unit 100 continues the process. Furthermore, if the out-of-range stay time B is exceeded before time T12 is reached, the control unit 100 stops the process.

The control unit 100 then performs a second process. The control unit 100 acquires the point at which the sensor value transitions from the upper limit (peak) to the lower limit (valley). The control unit 100 acquires the time T14 at which the sensor value becomes the lower limit threshold Amin. Here, if the interval R12 between time T12 and time T14 is within the range stay time C, the control unit 100 continues the process. Furthermore, if the range stay time C is exceeded before time T14 is reached, the control unit 100 stops the process.

That is, the control unit 100 repeatedly executes the first process and the second process. The control unit 100 compares the interval R14 between time T14 and time T16 with the out-of-range staying time B. The control unit 100 also compares the interval R16 between time T16 and time T18 with the in-range staying time C.

Then, when the time period from time T10, when the specific vibration is detected, exceeds the detection determination value D, the control unit 100 determines that the detected vibration is a specific vibration. This specific vibration is determined to be up to the end of the processing immediately before the end of the processing.

For example, suppose that the section R20 from time T20 to time T22 exceeds the out-of-range staying time B. In this case, the control unit 100 ends this process and sets the vibrations that have been detected up to that point as specific vibrations. At this time, the section of the specific vibration is set to be from T10, when the start was detected, to the time T20, when the process was most recently terminated.

Between time T24 and time T20, the control unit 100 executes the second process. The control unit 100 does not determine whether to end the process by comparing the time spent within range C with the second process. Therefore, time T20 is a valid time (range).

However, between time T20 and time T22, the control unit 100 executes the first process. The control unit 100 determines to end the specific vibration detection process because the out-of-range residence time B has been exceeded by the first process. Therefore, the control unit 100 detects the period up to the point at which the most recent process ended (time T20 at which the most recent second process ended) as a section containing specific vibration.

[4. Scratching behavior determination process]
[4.1 Processing flow]
A case where a specific movement/behavior is determined using the specific vibration detection process will be described below. Fig. 7 is an example illustrating the operation of a scratching behavior determination process for determining a scratching behavior as a specific behavior of the subject.

The control unit 100 executes a specific vibration detection process on the data (sensor value) acquired from the detection device 3. At this time, the control unit 100 may execute the process on multiple parameters in parallel.

For example, in FIG. 7, the control unit 100 sets the parameters to setting 1 and executes the specific vibration detection process (step S202 → step S204). The control unit 100 also sets the parameters to setting 2 and executes the specific vibration detection process (step S206 → step S208). The control unit 100 also sets the parameters to setting 3 and executes the specific vibration detection process (step S210 → step S212). In this way, the control unit 100 may set multiple parameters and execute multiple specific vibration detection processes. Note that, in FIG. 7, three settings are described as an example when the parameter table 156 shown in FIG. 3 is used, but more than this number may be used. The control unit 100 may also execute only one.

The control unit 100 ORs the sections (specific vibration sections) in which the specific vibrations output by the multiple executions of the specific vibration detection process were detected (step S214). The control unit 100 combines the sections determined to be specific vibrations output within the setting ranges of parameters 1, 2, and 3 into one section. The control unit 100 then determines that the section of the specific vibration is the section (time) in which the subject performed a scratching behavior (step S216).

The control unit 100 may output a report including the time period during which the scratching behavior was determined and the duration of the scratching behavior (step S218).

The control unit 100 may determine the scratching behavior time together with other information when calculating the time. For example, even if the specific vibration section is included, the control unit 100 may determine that the body movement before and after the determination that the subject got out of bed was associated with getting out of bed, and may determine that there was no scratching behavior. The control unit 100 may also calculate the scratching behavior time together with the sleep state when calculating the scratching behavior time. For example, the control unit 100 may calculate the scratching behavior time separately for when the subject was asleep and when awake. The control unit 100 may also output the presence or absence of scratching behavior together with the sleep diary.

[4.2 Operation example]
FIG. 8 is a diagram showing a sensor value in a waveform. In this embodiment, the detection device 3 acquires vibrations including the subject's body movement. For example, the detection device 3 outputs a sensor value including the subject's biological information and load changes due to movements such as turning over in bed. Therefore, the sensor value output from the detection device 3 is output in various ranges. For example, in FIG. 8(a), the sensor value is output almost in the center of the range that can be output (processed) as the sensor value. However, in FIG. 8(b), many sensor values are output at the lower limit side of the range that can be output (processed) as the sensor value. In addition, in FIG. 8(c), many sensor values are output at the upper limit side of the range that can be output (processed) as the sensor value. In addition, as is clear from the diagrams of FIG. 8(b) and FIG. 8(c), the sensor value includes a considerable portion that is out of the range that can be output (processed). In such a case, the control unit 100 cannot correctly detect a specific vibration based on one set range. Therefore, the control unit 100 of this embodiment detects a specific vibration based on different parameters in settings 1, 2, and 3.

As a result, for example in FIG. 8(a), a specific vibration can be correctly detected by using the parameters of setting 1. For example, it is determined that there is a possibility of a specific vibration from time T100, and a valid waveform (sensor value) is detected during interval R100 indicating detection determination value D. Then, an invalid waveform (sensor value) is detected after interval R105 has elapsed from time T105. Therefore, it is possible to detect that there is an interval of a specific vibration between time T100 and time T105, and the control unit 100 determines that the subject engaged in scratching behavior during this interval.

Similarly, in FIG. 8(b), a specific vibration can be correctly detected by using the parameters of setting 2. For example, it is determined that there is a possibility of a specific vibration from time T110, and a valid waveform (sensor value) is detected during section R110 indicating detection determination value D. Then, an invalid waveform (sensor value) is detected after section R115 has elapsed from time T115. Therefore, it is possible to detect that there is a section of a specific vibration between time T110 and time T115, and the control unit 100 determines that the subject engaged in a scratching behavior during this section.

In addition, in FIG. 8(c), the specific vibration can be correctly detected by using the parameters of setting 3. For example, it is determined that there is a possibility of a specific vibration from time T120, and a valid waveform (sensor value) is detected during section R120 indicating detection determination value D. Then, an invalid waveform (sensor value) is detected after section R125 has elapsed from time T125. Therefore, it is possible to detect that there is a section of a specific vibration between time T120 and time T125, and the control unit 100 determines that the subject engaged in a scratching behavior during this section.

The control unit 100 can then combine the results of these determinations to determine the subject's scratching behavior over a specified range (e.g., while sleeping, while in bed, or throughout the night).

Figure 9 is an example of a graph showing the output of actual sensor values. Figure 9 shows normal breathing and heart rate waveforms, and Figure 10 shows the waveform during scratching behavior. For example, in the case of Figure 9, the sensor values output from two sensor devices provided in the detection device 3 are shown above and below, and a large-cycle breathing waveform is displayed as a waveform that includes a heart rate waveform and some body vibration noise. In the case of Figure 10, compared to the normal breathing and heart rate waveforms of Figure 9, a waveform that further includes vibrations during scratching behavior is shown above and below.

Figure 10 also shows the location where the scratching behavior was performed based on a specific vibration determined by each parameter for the sensor value. For example, the area of the scratching behavior determined based on parameter 1 is G104, the area of the scratching behavior determined based on parameter 2 is G100, and the area of the scratching behavior determined based on parameter 3 is G102. In this way, by combining the areas of the scratching behavior determined by each parameter, the control unit 100 can more accurately determine the scratching behavior.

Furthermore, for example, the area of scratching behavior determined by the upper sensor device may be combined with the area of scratching behavior determined by the lower sensor device.

Figure 11 shows the results of estimating the scratching behavior of an atopic dermatitis patient overnight using the processing of this embodiment and the results of measuring the behavior using an infrared video camera, and shows the results of estimating the scratching behavior using a conventional method of activity amount. The graph G200 shows the locations where scratching behavior was determined to have occurred using the processing of this embodiment. The graph G210 shows the locations where scratching behavior of the subject was determined to have occurred by visual inspection by a human using an infrared video camera, as is conventionally done. The graph G220 shows the locations where scratching behavior of the subject was determined to have occurred based on sleep/wakefulness determined using a body vibration meter and activity amount.

The line graph shows the results of the sleep/wake assessment, with 0 on the second axis (right side) indicating an out-of-bed state, 1 indicating an awake (in bed) state, and 2 indicating a sleep state. The histogram indicates the amount of activity, calculated as a numerical value of body movement per minute. As the amount of activity includes not only scratching behavior but also body movements such as the subject turning over in their sleep, it is difficult to accurately estimate only scratching. This tendency is particularly pronounced for subjects who scratch less, as body movements other than scratching are observed relatively more frequently.

Figure 12 is a scatter plot showing the correspondence between the scratching time determined by video assessment and the scratching time obtained by calculation. Figure 12(a) is a scatter plot with the scratching time estimated from the activity level per hour on the vertical axis and the scratching time determined by video assessment on the horizontal axis. Figure 12(b) is a Brand-Altman plot with the average scratching time determined by video assessment and the estimated scratching time based on the activity level on the horizontal axis and the difference between the scratching time determined by video assessment and the estimated scratching time based on the activity level on the vertical axis.

Figure 12(c) is a scatter plot with the scratching time estimated from the specific vibration every hour on the vertical axis and the scratching time assessed by video on the horizontal axis. Figure 12(d) is a Brand-Altman plot with the average scratching time assessed by video and the estimated scratching time due to specific vibration on the horizontal axis and the difference between the scratching time assessed by video and the estimated scratching time due to specific vibration on the vertical axis. The white circles (○) on the graph show the measurement results of patients with atopic dermatitis, and the crosses (×) show the measurement results of healthy subjects. Healthy subjects do not scratch, so the scratching time assessed by video is 0.

Comparing Figures 12(a) and (c), it can be seen that when estimating scratching time based on the amount of activity in the conventional method in Figure 12(a), particularly in healthy subjects or in areas where the amount of scratching per hour is small, the estimation method based on the amount of activity judges the scratching time to be longer, whereas the estimation method based on specific vibration in Figure 12(c) shows an improvement and obtains a correct correlation. Also, comparing Figures 12(b) and 12(d) it is clear that the average value and variance of the error from the video judgment are smaller in this embodiment, and the scratching time can be identified effectively.

5. Modifications
Although an embodiment of the present invention has been described in detail above with reference to the drawings, the specific configuration is not limited to this embodiment, and designs that do not deviate from the gist of the present invention are also included in the scope of the claims.

For example, a frequency analysis such as a fast Fourier transform (FFT) can be performed on vibration data from the past 30 seconds or so that can be acquired by the detection device 3, and a specific vibration can be determined when a peak is detected in the frequency and the amplitude of the waveform exceeds a set range. In this case, since a vibration having a frequency at which a peak is detected has been observed, it can be determined that a periodic vibration has been detected.

The frequency at which the peaks are detected is, for example, as described in the above embodiment, and the peaks are detected at a frequency higher than respiration (1 Hz or higher). The setting range of the waveform amplitude is, for example, stored in parameter table 156, and refers to the setting range of the upper threshold Amax (for example, "2458") and the lower threshold Amin (for example, "1638").

Figure 13 shows the acquired waveform and a power spectrum that is a graph of the power (vertical axis) per frequency (horizontal axis) of the acquired waveform.

Here, Fig. 13(a) and (b) show normal respiratory heart rate waveforms, Fig. 13(c) and (d) show normal body movement waveforms such as rolling over, and Fig. 13(e) and (f) show specific vibration waveforms during scratching behavior. Fig. 13(a), (c) and (e) show waveforms obtained before application, and Fig. 13(b), (d) and (f) show power spectra graphed by power (vertical axis) for each frequency (horizontal axis).

In a normal respiratory heart rate waveform, a power spectrum peak (vibration due to breathing) is observed below 1 Hz, but in a specific vibration waveform, a power spectrum peak is observed in the 1 to 3 Hz range, and no clear peak is observed in the body movement waveform.

When the duration of a particular vibration is short, the spectral peak may be buried in other vibrations, so it is advisable to use at least 10 seconds of data to capture the respiratory waveform with this method.

In addition, in this embodiment, the processing device 5 outputs the biometric information based on the results output by the detection device 3, but the detection device 3 may calculate everything. Also, instead of implementing this by installing an application on a terminal device (e.g., a smartphone, tablet, or computer), for example, the processing may be performed on the server side and the processing results may be returned to the terminal device.

For example, the detection device 3 may upload vibration data to a server, thereby realizing the above-mentioned processing on the server side. The detection device 3 may be realized by a device such as a smartphone with a built-in acceleration sensor and vibration sensor.

In the above-described embodiment, the specific state (movement) of the subject is determined to be a scratching movement, but other states or movements that can be determined from high-frequency vibrations may also be used. For example, it is possible to determine whether the subject is having convulsions, or is experiencing tremors or shaking.

In addition, in the embodiments, the programs that run on each device are programs that control the CPU and the like (programs that make the computer function) so as to realize the functions of the above-described embodiments. Information handled by these devices is temporarily stored in a temporary storage device (e.g., RAM) during processing, and is then stored in various ROM, HDD, and SSD storage devices, and is read, modified, and written by the CPU as necessary.

When distributing the program on the market, the program can be stored on a portable recording medium and distributed, or transferred to a server computer connected via a network such as the Internet. In this case, the storage device of the server computer is of course included in the present invention.

REFERENCE SIGNS LIST 1 System 3 Detection device 3 Processing device 100 Control unit 102 Specific vibration detection unit 104 Biometric information calculation unit 106 Subject state determination unit 120 Signal acquisition unit 150 Storage unit 152 Subject data storage area 154 Sensor value storage area 156 Parameter table 160 Input unit 170 Output unit 10 Bed 20 Mattress

Claims (7)

The bed device includes an acquisition unit that acquires a sensor value indicating vibration from a sensor placed between the bed device and a subject, and a control unit,
The control unit is
Detecting a specific vibration that is higher in frequency than respiration vibration, higher in amplitude than heartbeat vibration, and periodic from the sensor value;
A determination device that determines that a section in which the specific vibration is detected is a section in which the subject performs a specific behavior,
The control unit is
A setting range between an upper threshold value and a lower threshold value of the sensor value is set,
the time during which the sensor value is greater than an upper threshold value or less than a lower threshold value of the set range is within a first determination value,
Furthermore, it is determined that the specific vibration is detected during a period during which the sensor value transitions from a value greater than an upper threshold value of the set range to a value less than a lower threshold value, or during which the sensor value transitions from a value less than the lower threshold value of the set range to a value greater than an upper threshold value, within a second determination value.
A determination device characterized by: The control unit is
the time during which the sensor value is greater than an upper threshold value or less than a lower threshold value of the set range exceeds a first determination value,
Furthermore, when a time taken for the sensor value to transition from a value greater than an upper threshold value of the set range to a value less than a lower threshold value, or a time taken for the sensor value to transition from a value less than the lower threshold value of the set range to a value greater than an upper threshold value, exceeds a second determination value, it is determined that the section in which the specific vibration is detected has ended.
The determination device according to claim 1 . The bed device includes an acquisition unit that acquires a sensor value indicating vibration from a sensor placed between the bed device and a subject, and a control unit,
The control unit is
Detecting a specific vibration that is higher in frequency than respiration vibration, higher in amplitude than heartbeat vibration, and periodic from the sensor value;
A determination device that determines that a section in which the specific vibration is detected is a section in which the subject performs a specific behavior,
The control unit is
A setting range between an upper threshold value and a lower threshold value of the sensor value is set,
The specific vibration section can be detected as the specific vibration after a determination time has elapsed since the vibration first exceeded the set range,
a period during which the sensor value is greater than an upper threshold value or less than a lower threshold value of the set range is within a first determination value, the period being determined as a period during which the specific vibration is detected;
A determination device characterized by: The bed device includes an acquisition unit that acquires a sensor value indicating vibration from a sensor placed between the bed device and a subject, and a control unit,
The control unit is
Detecting a specific vibration that is higher in frequency than respiration vibration, higher in amplitude than heartbeat vibration, and periodic from the sensor value;
A determination device that determines that a section in which the specific vibration is detected is a section in which the subject performs a specific behavior,
The frequency higher than the vibration caused by breathing means that when the vibration (body movement) caused by breathing is 0.1 to 0.6 Hz, the specific vibration is 1 Hz or higher.
The amplitude higher than the vibration caused by the heartbeat is equal to or greater than 1/5 of the maximum amplitude caused by breathing or equal to or greater than 1/5 of the measurement range.
A determination device characterized by: A computer capable of acquiring a sensor value indicating vibration from a sensor placed between the bed apparatus and the subject,
A function of detecting a specific vibration having a higher frequency than a vibration caused by breathing, a higher amplitude than a vibration caused by a heartbeat, and a periodicity from the sensor value;
A determination function that determines that the section in which the specific vibration is detected is a section in which the subject performed a specific behavior;
A program capable of realizing the above,
The determination function is
A setting range between an upper threshold value and a lower threshold value of the sensor value is set,
the time during which the sensor value is greater than an upper threshold value or less than a lower threshold value of the set range is within a first determination value,
Furthermore, it is determined that the specific vibration is detected during a period during which the sensor value transitions from a value greater than an upper threshold value of the set range to a value less than a lower threshold value, or during which the sensor value transitions from a value less than the lower threshold value of the set range to a value greater than an upper threshold value, within a second determination value.
Program. A computer capable of acquiring a sensor value indicating vibration from a sensor placed between the bed apparatus and the subject,
A function of detecting a specific vibration having a higher frequency than a vibration caused by breathing, a higher amplitude than a vibration caused by a heartbeat, and a periodicity from the sensor value;
A determination function that determines that the section in which the specific vibration is detected is a section in which the subject performed a specific behavior;
A program capable of realizing the above,
The determination function is
A setting range between an upper threshold value and a lower threshold value of the sensor value is set,
The specific vibration section can be detected as the specific vibration after a determination time has elapsed since the vibration first exceeded the set range,
a period during which the sensor value is greater than an upper threshold value or less than a lower threshold value of the set range is within a first determination value, the period being determined as a period during which the specific vibration is detected;
Program. A computer capable of acquiring a sensor value indicating vibration from a sensor placed between the bed apparatus and the subject,
A function of detecting a specific vibration having a higher frequency than a vibration caused by breathing, a higher amplitude than a vibration caused by a heartbeat, and a periodicity from the sensor value;
A determination function that determines that the section in which the specific vibration is detected is a section in which the subject performed a specific behavior;
A program capable of realizing the above,
The frequency higher than the vibration caused by breathing means that when the vibration (body movement) caused by breathing is 0.1 to 0.6 Hz, the specific vibration is 1 Hz or higher.
The amplitude higher than the vibration caused by the heartbeat is equal to or greater than 1/5 of the maximum amplitude caused by breathing or equal to or greater than 1/5 of the measurement range.
Program.

Images