Relaxing state determination system and relaxing state determination method
The system uses muscle displacement sensors and machine learning to accurately determine relaxation states, addressing inaccuracies in existing methods by employing threshold comparisons and machine learning for precise relaxation assessment.
Patent Information
- Application Number
- JP2023085150
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-05-24
- Publication Date
- 2024-12-05
AI Technical Summary
Existing methods for determining a relaxed state, such as those based on facial expressions and heart rate or brain waves, suffer from inaccuracies and require complex setups, making it difficult to detect relaxation with high precision.
A system utilizing muscle displacement detection sensors, such as calf and trapezius muscle sensors, to determine relaxation states by analyzing muscle displacement data, combined with threshold comparisons and machine learning algorithms, to accurately assess whether a subject is in a relaxed or tense state.
Enables precise determination of relaxation states with high accuracy and simplicity, using muscle displacement data from sensors like calf and trapezius muscle sensors, improving detection accuracy.
Smart Images

Figure 2024168450000001_ABST
Abstract
Description
[Technical field]
[0001] The present invention relates to a relaxation state determination system and a relaxation state determination method for determining whether a subject is in a relaxed state. [Background technology]
[0002] Conventionally, various methods for measuring mental and physical health conditions have been proposed. For example, Patent Document 1 proposes a technique for determining a subject's relaxation state based on the facial expression and heart rate detected from a facial image of the subject. Patent Document 2 also proposes a technique for determining a subject's relaxation state based on electroencephalograms. [Prior art documents] [Patent documents]
[0003] [Patent Document 1] JP 2022-133877 A [Patent Document 2] JP 2023-11578 A Summary of the Invention [Problem to be solved by the invention]
[0004] However, changes in facial expressions and heart rate vary greatly from person to person, and even if a value indicating a state of relaxation is detected by detecting a facial image or heart rate using the technology described in Patent Document 1, the detected value is relative, making it difficult to detect the degree of relaxation with high accuracy. Also, as described in Patent Document 2, it is possible to detect the degree of relaxation from brain waves, but a complex configuration is required to detect and analyze brain waves, making it difficult to accurately determine the state of relaxation with a simple configuration.
[0005] An object of the present invention is to provide a relaxation state determination system and a relaxation state determination method that can determine a relaxation state easily with high accuracy. [Means for solving the problem]
[0006] The relaxed state determination system of the present invention comprises a muscle displacement detection sensor that detects muscle displacement at a specific location of a subject, and a relaxed state determination unit that determines whether the subject is in a relaxed state or a tense state based on the muscle displacement detected by the muscle displacement detection sensor.
[0007] In addition, the relaxed state determination method of the present invention includes a muscle displacement detection process that detects muscle displacement at a specific location of the subject, and a relaxed state determination process that determines whether the subject is in a relaxed state or a tense state based on the muscle displacement detected by the muscle displacement detection process. Effect of the Invention
[0008] According to the present invention, it is possible to easily and appropriately determine whether a person is in a relaxed state or a tense state. [Brief description of the drawings]
[0009] [Figure 1] 1 is a configuration diagram showing an example of a relaxation state determination system according to an embodiment of the present invention; [Diagram 2] 1 is a diagram showing an example of mounting a sensor device used in a relaxation state determination system according to an embodiment of the present invention; [Diagram 3] 1 is a perspective view showing an example (front surface) of a sensor device used in a relaxation state determination system according to an embodiment of the present invention. FIG. [Figure 4] 1 is a perspective view showing an example (rear surface) of a sensor device used in a relaxation state determination system according to an embodiment of the present invention. FIG. [Diagram 5] 5 is a flowchart showing an example of initial settings in the relaxation state determination system according to the embodiment of the present invention. [Figure 6] FIG. 1 is a diagram showing an example of a screen at the time of initial setup in a relaxation state determination system according to an embodiment of the present invention. [Figure 7]5 is a flowchart showing an example of a determination process (an example of threshold comparison) performed by the relaxed state determination system according to an embodiment of the present invention. [Figure 8] 1 is a flowchart showing an example of a determination process (an example of machine learning) performed by a relaxation state determination system according to an embodiment of the present invention. [Figure 9] FIG. 13 is a diagram showing an example of a display of a determination result obtained by the relaxation state determination system according to the embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0010] Hereinafter, a relaxation state determination system and a relaxation state determination method according to one embodiment of the present invention (hereinafter referred to as "this embodiment") will be described with reference to the accompanying drawings.
[0011] [Configuration of the relaxation state determination system] 1, the relaxed state determination system of this example is composed of a calf sensor 10 or a trapezius muscle sensor 20 attached to the body B of a subject, and a terminal 100 that determines and displays the relaxed state based on the muscle displacement of a muscle (calf muscle or trapezius muscle) detected by the calf sensor 10 or trapezius muscle sensor 20. The calf sensor 10 and the trapezius muscle sensor 20 are muscle displacement detection sensors that detect muscle displacement at the location where they are attached, and it is sufficient to attach either the calf sensor 10 or the trapezius muscle sensor 20 to the body B.
[0012] A smartphone, a smart watch, or the like can be applied as the terminal 100. Configuration examples of the calf sensor 10 and the trapezius muscle sensor 20 will be described later. The calf sensor 10 or the trapezius muscle sensor 20 is a sensor that detects muscle displacement in multiple channels, and supplies the detected signals, that is, muscle displacement data, to the terminal 100 by wired or wireless transmission.
[0013] 1, the terminal 100 is configured with a hardware configuration including a CPU (Central Processing Unit) 111, a memory 112, a sensor interface 113, a display unit 106, and a communication interface 107. The terminal 100 realizes each processing unit described below by executing an application program implemented in the memory 112 by the CPU 111.
[0014] First, a description will be given of the configuration in terms of functions executed by the terminal 100. The terminal 100 includes a muscle displacement data acquisition unit 101, a threshold determination unit 102, a machine learning unit 103, an initial setting unit 104, a relaxed state determination unit 105, and a display unit 106. The muscle displacement data acquisition unit 101 acquires muscle displacement data from the calf sensor 10 or the trapezius muscle sensor 20. Here, muscle displacement data of 14 channels is acquired. The threshold determination section 102 compares the muscle displacement data of the 14 channels with a preset threshold, and supplies the number of channels for which it has been determined that the muscle displacement data has displaced by more than the threshold to the relaxed state determination section 105 .
[0015] The machine learning unit 103 classifies the muscle displacement data of the 14 channels by machine learning processing, and classifies the muscle displacement data of each channel as being in a tense state or a relaxed state. The machine learning unit 103 then supplies the classified data to the relaxed state determination unit 105. The initial setting unit 104 executes initial setting when the calf sensor 10 or the trapezius muscle sensor 20 is attached to the body B of the subject. The relaxed state determination unit 105 determines whether the subject is in a relaxed state or a tense state based on either or both of the output data of the threshold determination unit 102 and the output data of the machine learning unit 103.
[0016] The display unit 106 displays the result of the determination made by the relaxed state determination unit 105 . The result of the determination by the relaxation state determination section 105 may be transmitted to a server or another terminal via the Internet or the like. In addition, in FIG. 1, the terminal 100 is configured to include both the threshold determination unit 102 and the machine learning unit 103, but the terminal 100 may be configured to include only one of them.
[0017] [Sensor configuration and installation example] FIG. 2 shows an example in which the calf sensor 10 is attached to the calf of one leg f of a body B. As shown in FIG. 2, the calf sensor 10 is attached so as to detect, for example, the displacement (tension) of a muscle (gastrocnemius) slightly above the ankle (for example, about 10 cm above) at the back as muscle displacement data.
[0018] 3 and 4 show examples of the configuration of the calf sensor 10 that is attached as shown in FIG. FIG. 3 shows the front side of the calf sensor 10, and FIG. The calf sensor 10 is composed of a main body 11 and a belt 12. The belt 12 is made of a flexible resin such as silicone rubber. By joining one end 12a and the other end 12b of the belt 12, the calf sensor 10 is attached around the calf as shown in FIG. The main body 11 of the calf sensor 10 includes a built-in circuit board for detecting muscle displacement and a battery.
[0019] 4, 14 sensor members 13 are successively arranged on the back side of the calf sensor 10 along the longitudinal direction of the belt 12. A lens is arranged on each sensor member 13, and the displacement of the calf muscle (gastrocnemius muscle) is detected by these lenses. For example, an optical distance sensor that detects distance using infrared rays is used as the sensor member 13, and each sensor member 13 performs muscle displacement detection processing based on the detected distance. When the calf sensor 10 is attached, the 14 sensor members 13 are arranged behind the calves. The back of the calves is an area where muscles are concentrated.
[0020] The 14-channel muscle displacement data detected by the 14 sensor members 13 is digitized by a circuit in the main body 11 and then wirelessly transmitted to the terminal 100. Alternatively, the calf sensors 10 and the terminal 100 may be connected by a wire, and the muscle displacement data may be transmitted to the terminal 100 by the wire.
[0021] Although the configuration of the trapezius muscle sensor 20 is not shown in the figures, the trapezius muscle sensor 20 is configured to detect muscle displacement using the same principle as the calf sensor 10. That is, the trapezius muscle sensor 20 is attached to a location from the neck to the upper back of the subject's body B. The area from the neck to the upper back of the subject's body B is where the trapezius muscles are concentrated. The trapezius muscle sensor 20 detects the displacement of the trapezius muscle on 14 channels by arranging 14 sensor members similar to the sensor members 13 of the calf sensor 10 shown in Fig. 4 on its attachment surface. Note that, like the sensor members 13 of the calf sensor 10, the detected muscle displacement data of the trapezius muscle is transmitted to the terminal 100 wirelessly or via wire.
[0022] [Initial setup process] When determining a relaxed state using the relaxed state determination system of this example, initial settings are required in the terminal 100 after the calf sensor 10 or the trapezius muscle sensor 20 is attached to the subject's body B. As already described in FIG. 1, the initial settings are performed by the initial setting unit 104.
[0023] FIG. 5 is a flowchart showing a flow of the initial setting by the initial setting unit 104. First, when starting the initial setting, the initial setting unit 104 instructs the subject to exert force by displaying it on the display unit 106 (step S11). Here, for example, when the calf sensor 10 is used, the subject is instructed to exert force on the foot, and when the trapezius muscle sensor 20 is used, the subject is instructed to exert force on the shoulder.
[0024] After issuing the instruction in step S11, the initial setting unit 104 learns the muscle displacement data detected by the calf sensor 10 or the trapezius muscle sensor 20 as muscle displacement data during tension (step S12). In this step S12, the initial setting unit 104 calculates the maximum value, average value, etc. of the muscle displacement data of, for example, 14 channels to set the muscle displacement data during tension. Here, even when the average value of the muscle displacement data during tension is used, it may be the average value of a predetermined number of channels at the top, and does not necessarily have to be a simple average of the muscle displacement data of all 14 channels. Alternatively, when using the muscle displacement data during tension for machine learning, the values of all muscle displacement data of 14 channels may be used.
[0025] Next, the initial setting unit 104 instructs the subject to relax by displaying on the display unit 106 (step S13). For example, when the calf sensor 10 is used, the subject is instructed to relax the legs, and when the trapezius muscle sensor 20 is used, the subject is instructed to relax the shoulders. After issuing the instruction in step S13, the initial setting unit 104 learns the muscle displacement data detected by the calf sensor 10 or the trapezius muscle sensor 20 as muscle displacement data in a relaxed state (step S14). Here too, the initial setting unit 104 calculates the maximum value, average value, etc. of the muscle displacement data of, for example, 14 channels to set it as the muscle displacement data in the relaxed state. Alternatively, all muscle displacement values of the 14 channels may be used as the muscle displacement data in the relaxed state for machine learning.
[0026] Then, the initial setting unit 104 sets the threshold determination unit 102 and the machine learning unit 103 based on the values of the muscle displacement data during tension learned in step S12 and the values of the muscle displacement data during relaxation learned in step S13 (step S15). That is, the threshold value determination unit 102 sets a determination threshold value TH1 between the value of muscle displacement data when tense and the value of muscle displacement when relaxed. Further, the machine learning unit 103 is set to use the obtained values of muscle displacement data during tension and muscle displacement data during relaxation as teacher signals for machine learning.
[0027] FIG. 6 shows an example of the display screen of the display unit 106 at the time of initial setting. The example in FIG. 6 shows an instruction screen in a state where the calf sensor 10 is used to instruct the subject to exert force in step S11. At this time, the screen in FIG. 6 displays a message 201 saying "Pull your foot towards you and keep your calf muscles tense," along with an image 202 showing the position of the foot at that time.
[0028] The screen in FIG. 6 also displays a measurement in progress button 203 and an end button 204 so that the subject can touch to start or end the measurement. When instructing the subject to relax in step S13, the instruction to relax is also displayed on a similar screen. After the above-described initial settings are completed, the process of determining the relaxed state in the relaxed state determination system of this embodiment is started.
[0029] [Relaxed state determination processing using threshold judgment] FIG. 7 is a flowchart showing a process in which the relaxed state determination section 105 determines the relaxed state based on the determination result in the threshold determination section 102. First, the muscle displacement data acquiring unit 101 acquires 14-channel muscle displacement data, and supplies the acquired 14-channel muscle displacement data to the threshold determination unit 102 (step S21). The threshold value determination unit 102 compares the values of the acquired muscle displacement data of the 14 channels with the threshold value TH1 set in the initial setting process (step S22).
[0030] Then, the threshold value determining unit 102 determines whether or not more than half of the 14 channels, that is, 7 or more channels, meet the threshold value TH1 as a result of the comparison in step S22 (step S23). In step S23, if more than half of the channels are equal to or greater than the threshold TH1 (Yes in step S23), the relaxed state determination unit 105 determines that the subject is in a tense state (step S24).In addition, in step S23, if more than half of the channels are not equal to or greater than the threshold TH1 (No in step S23), the relaxed state determination unit 105 determines that the subject is in a relaxed state (step S25).
[0031] Then, the relaxed state determination unit 105 causes the display unit 106 to display a display screen based on the determination result (step S26). For example, the display unit 106 displays "You are in a relaxed state" or "You are in a tense state" based on the determination result. In addition, the relaxed state determination section 105 may transmit the determination result to a server or the like via a communication section of the terminal 100, and the server may register whether the subject is in a relaxed state or a tense state.
[0032] [Relaxation state determination processing using machine learning] FIG. 8 is a flowchart showing a process in which the relaxed state determination section 105 determines the relaxed state based on the classification result in the machine learning section 103. First, the muscle displacement data acquisition unit 101 acquires muscle displacement data of 14 channels, and supplies the acquired 14 channels to the machine learning unit 103 (step S31). The machine learning unit 103 classifies the acquired values of the 14-channel muscle displacement data using the teacher signal obtained in the initial setting (step S32).
[0033] Then, the relaxed state determination section 105 acquires the result of classification in step S32, and determines whether or not the classification result is classified as a tense state (step S33). In step S33, if the classification result is a tense state (Yes in step S33), the relaxed state determination unit 105 determines that the subject is in a tense state (step S34). In addition, in step S33, if the subject is not in a tense state (No in step S33), the relaxed state determination unit 105 determines that the subject is in a relaxed state (step S35).
[0034] Then, the relaxed state assessment section 105 causes a display screen based on the assessment result to be displayed on the display section 106 (step S36). Also in this case, the relaxed state assessment section 105 may transmit the assessment result to a server or the like via a communication section of the terminal 100, and the server side may register whether the subject is in a relaxed state or a tense state.
[0035] [Example of a relaxed state] FIG. 9 shows an example of the display screen of the terminal 100 displaying a relaxed state and a tense state. The example of Figure 9 is an example in which terminal 100 displays a relaxed state or a tense state obtained from a calf sensor 10 or a trapezius muscle sensor 20 connected to the terminal 100 itself, as well as a relaxed state or a tense state obtained from another terminal. 9, the states of the measurement subjects A1 and A2 are displayed as face images. In this example, a rain mark M1 is displayed beside the face of the measurement subject A1, and the face is colored dark to indicate that the person is in a tense state. Also, a sunny mark M2 is displayed beside the face of the measurement subject A2, and the face is colored bright to indicate that the person is in a relaxed state.
[0036] As described above, the relaxation state determination system of this embodiment can accurately determine and display whether a subject is in a relaxed state or a tense state from muscle displacement data of the calf or trapezius muscles. In particular, by making a determination from the calf or trapezius muscles, the detection accuracy can be significantly improved.
[0037] [Variations] It should be noted that the embodiment examples described so far have been described in detail in order to explain the present invention in an easily understandable manner, and various modifications and changes are possible to the configurations and processes described in the above-mentioned embodiment examples. For example, in this example, 14 channels of muscle displacement data are acquired from the calf sensor 10 or the trapezius muscle sensor 20, and the relaxed state is determined from the 14 channels of muscle displacement data, but the relaxed state may be determined from muscle displacement data of a fewer number of channels. However, since the threshold determination unit 102 performs majority decision to determine whether more than half of the channels are equal to or greater than the threshold in step S23 (FIG. 7), it is preferable to acquire muscle displacement data of at least three channels, but it may also be determined that the subject is tense when one of two channels of muscle displacement data exceeds the threshold. Alternatively, it may be determined that the subject is tense when one channel of muscle displacement data exceeds the threshold.
[0038] In addition, it is preferable to perform the initial settings when the sensor 10 or 20 is attached, but after the initial settings are performed once, the same subject may be judged using the data obtained in the initial settings even after the sensor 10 or 20 is removed. In addition, the teacher signal for machine learning may be obtained from the settings of many subjects to improve the judgment accuracy.
[0039] 1, the relaxed state determination system uses a terminal 100 such as a smartphone, but the calf sensor 10 or the trapezius muscle sensor 20 may have a built-in processing unit such as a microcomputer that performs the determination process shown in the flowchart of Fig. 7 or 8. Alternatively, muscle displacement data obtained by the calf sensor 10 or the trapezius muscle sensor 20 may be transmitted to an external server, and the relaxed state or tense state may be determined on the server side. In this case, the server side will implement a program that performs the determination process shown in the flowchart of Fig. 7 or 8. [Explanation of symbols]
[0040] 10...calf sensor, 11...main body, 12...belt, 13...sensor member, 20...trapezius muscle sensor, 100...terminal, 101...muscle displacement data acquisition unit, 102...threshold determination unit, 103...machine learning unit, 104...initial setting unit, 105...relaxation state determination unit, 106...display unit, 107...communication interface, 111...CPU, 112...memory, 113...sensor interface, 201...message display, 202...image display, 203...measurement in progress button, 204...end button
Claims
1. A muscle displacement detection sensor that detects muscle displacement at a specific location of a subject; and a relaxation state determination unit that determines whether the subject is in a relaxed state or a tense state based on the muscle displacement detected by the muscle displacement detection sensor. Relaxation state determination system.
2. The present invention further includes a threshold determination unit that compares the value of the muscle displacement detected by the muscle displacement detection sensor with a predetermined threshold value, The relaxation state determination unit determines that the subject is relaxed when the muscle displacement value determined by the threshold determination unit is less than the threshold, and determines that the subject is tense when the muscle displacement value exceeds the threshold. The relaxation state determination system according to claim 1 .
3. The muscle displacement detection sensor detects muscle displacement at the specific location in N channels (N is 1 or more), The threshold determination unit determines that the subject is relaxed when the muscle displacement values of at least half of the N channels are less than the predetermined threshold, and determines that the subject is tense when the muscle displacement values of at least half of the channels exceed the threshold. The relaxation state determination system according to claim 2 .
4. As an initial setting, the muscle displacement value detected by the muscle displacement detection sensor is measured when the subject is instructed to exert force on the specific location, and the muscle displacement value detected by the muscle displacement detection sensor is measured when the subject is instructed to relax the specific location, and the threshold value is set based on the measured muscle displacement values. The relaxation state determination system according to claim 2 .
5. Further, a machine learning unit that performs machine learning based on the muscle displacement value detected by the muscle displacement detection sensor is provided. The relaxation state determination unit determines whether the subject is in a relaxed state or a tense state based on a result of the machine learning process performed by the machine learning unit. The relaxation state determination system according to claim 1 .
6. The muscle displacement detection sensor is a sensor that detects the displacement of the calf muscle. The relaxation state determination system according to claim 1 .
7. The muscle displacement detection sensor is a sensor that detects muscle displacement of the trapezius muscle. The relaxation state determination system according to claim 1 .
8. A muscle displacement detection process for detecting muscle displacement at a specific location of a subject; and a relaxation state determination process for determining whether the subject is in a relaxed state or a tense state based on the muscle displacement detected by the muscle displacement detection process. A method for determining a state of relaxation.
Citation Information
Patent Citations
Relaxing degree determination device, relaxing degree determination method, and relaxing guide device
JP2022133877A
Electroencephalogram measuring device, electroencephalogram measuring method, and electroencephalogram measuring program
JP2023011578A
