Biofeedback system, biofeedback method, and computer program

Abstract

To provide a biofeedback system which executes accurate output in response to a muscular activity.SOLUTION: A biofeedback system 100 is a biofeedback system which executes operation using a myoelectric signal input from a myoelectric sensor 30 for measuring myoelectricity of a user. The operation includes correcting a threshold stored in accordance with a measurement value obtained from the user, transforming the value of the myoelectric signal obtained from the myoelectric sensor to a pulse signal on the basis of comparison with the corrected threshold, and generating an output signal according to an exercise of the user by using the pulse signal.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to a biofeedback system, a biofeedback method, and a computer program. [Background technology]

[0002] For example, as disclosed in Japanese Patent Laid-Open No. 2011-56243 (hereinafter referred to as Patent Document 1), a biofeedback system using myoelectric signals is known. The myoelectric signals used in the biofeedback system are directly detected from a living body using electrodes placed on the skin surface. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2011-56243 A Summary of the Invention

[0004] However, the value of the myoelectric signal is prone to variations unrelated to exercise, such as the part where the electrodes are attached and the environment. For example, even with the same exercise load, the value of the myoelectric signal may differ depending on the part where the electrodes are attached. In addition, even with the same exercise load, the signal value may differ depending on the temperature and moisture content of the skin surface. For example, even with the same exercise load, the signal value may differ between the start and middle of exercise due to an increase in body temperature and sweating caused by exercise.

[0005] When an electromyogram is used in a biofeedback system, the electromyogram is subject to fluctuations unrelated to exercise, and the output signal may not correspond to the actual exercise load.

[0006] The present disclosure provides a biofeedback system, a biofeedback method, and a computer program that provide an output that is accurately in accordance with muscle activity.

[0007] According to one embodiment, the biofeedback system is a biofeedback system that performs a calculation using an electromyographic signal input from an electromyographic sensor that measures the electromyographic potential of a user, the calculation including: correcting a stored threshold value according to a measurement value obtained from the user; converting a value of the electromyographic signal obtained from the electromyographic sensor into a pulse signal based on a comparison with the corrected threshold value; and using the pulse signal to generate an output signal according to the user's movement.

[0008] According to another embodiment, a biofeedback method is a biofeedback method using an electromyogram signal input from an electromyogram sensor that measures a user's electromyogram, comprising: correcting a stored threshold value in accordance with a measurement value obtained from the user; converting a value of the electromyogram signal obtained from the electromyogram sensor into a pulse signal based on a comparison with the corrected threshold value; and using the pulse signal to generate an output signal in accordance with the user's movement.

[0009] According to another embodiment, the computer program causes a computer to perform a calculation using an electromyogram signal input from an electromyogram sensor that measures a user's electromyogram, the calculation including: correcting a stored threshold value according to a measurement value obtained from the user; converting a value of the electromyogram signal obtained from the electromyogram sensor into a pulse signal based on a comparison with the corrected threshold value; and using the pulse signal to generate an output signal according to the user's movement.

[0010] Further details will be described in the following embodiments. [Brief description of the drawings]

[0011] [Figure 1] FIG. 1 is a schematic diagram of a configuration of a biofeedback system according to an embodiment. [Diagram 2] FIG. 2 is a schematic block diagram showing the configuration of a computing device included in the biofeedback system. [Diagram 3] FIG. 3 is a diagram for explaining the pulse generation process. [Figure 4] FIG. 4 is a diagram for explaining the output conditions set for each sound. [Diagram 5] FIG. 5 is a flowchart showing a biofeedback method in the biofeedback system according to the first embodiment. [Figure 6] FIG. 6 is a diagram for explaining the sound recognition process in the arithmetic device according to the second embodiment. [Figure 7] FIG. 7 is a flowchart showing a biofeedback method in the biofeedback system according to the second embodiment. [Figure 8] FIG. 8 is a diagram for explaining the sound recognition process in the arithmetic device according to the third embodiment. [Figure 9] FIG. 9 is a flowchart showing a biofeedback method in the biofeedback system according to the third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] <1. Overview of Biofeedback System, Biofeedback Method, and Computer Program>

[0013] (1) A biofeedback system according to an embodiment is a biofeedback system that performs a calculation using an electromyographic signal input from an electromyographic sensor that measures a user's electromyographic activity, the calculation including: correcting a stored threshold value in accordance with a measurement value obtained from the user; converting the value of the electromyographic signal obtained from the electromyographic sensor into a pulse signal based on a comparison with the corrected threshold value; and using the pulse signal to generate an output signal in accordance with the user's movement.

[0014] Exercise refers to a movement accompanied by muscle activity, and the exercise load is measured by electromyography. The output is, for example, sound output, image output, vibration or other vibration, a combination of these, etc. Note that sound output includes human voice, i.e., sounds representing words, sounds without words such as bell sounds, and combinations of these.

[0015] The measured values ​​obtained from the user are values ​​detected from the user's body, such as the humidity and temperature of the skin surface. By correcting the stored threshold value according to the measured values ​​obtained from the user, the threshold value is corrected taking into account the user's physical condition.

[0016] Converting into a pulse signal based on a comparison with a threshold value means, for example, converting into a pulse signal in which the period during which the myoelectric signal value exceeds the threshold value is an on period and the period during which the myoelectric signal value does not exceed the threshold value is an off period. The threshold value can be set, for example, to a value equivalent to the value of the myoelectric signal when a specific exercise is performed. Note that the target exercise refers to the state of exercise that is targeted for a certain exercise. The state of exercise refers to the amount of exercise, posture during exercise, speed of movement, etc. The posture during exercise refers to the shape of the body during exercise, for example, the angle of the upper body while running, the angle of the knees while squatting, etc.

[0017] The EMG signal is detected using electrodes in contact with the skin. Therefore, it is easily affected by the humidity of the skin surface and movements unrelated to exercise. In particular, when detecting from a subject who is exercising, the environment of the skin surface is also easily changed by sweating and heat generation. Therefore, the fluctuation of the EMG signal value may become large. As a result, if an output signal is generated using features obtained from the EMG signal value, the generation of the output signal may be affected by the fluctuation of the EMG signal value.

[0018] In this regard, by correcting the threshold value taking into account the user's physical condition and converting the EMG signal into a pulse signal, the effect of fluctuations in the EMG signal value on calculations is reduced compared to using features obtained from the EMG signal value, and an output signal corresponding to the user's movement is generated.

[0019] (2) Preferably, correcting the stored threshold value includes correcting the stored threshold value using a measured value obtained from the user during measurement with the myoelectric sensor. Measurement with the myoelectric sensor is performed while the user is exercising, and there is a high possibility that the environment of the skin surface will change during the user's exercise, such as sweating or heat generation. Therefore, the threshold value is corrected taking into consideration the user's physical condition at the time of myoelectric detection. As a result, an output signal corresponding to the user's exercise with greater accuracy can be generated.

[0020] (3) Preferably, the threshold value is a predetermined value obtained from a time change in the value of the myoelectric signal when a prescribed motion is performed. The prescribed motion is, for example, a target motion. This makes it possible to generate an output signal according to the relationship between the user's motion and the prescribed motion.

[0021] (4) Preferably, the conversion to a pulse signal includes setting a period during which the waveform, which is a time change in the value of the myoelectric signal, exceeds a threshold to a constant value, thereby converting the myoelectric signal into a pulse signal in which the period during which the threshold is exceeded is an on-period and the period during which the threshold is not exceeded is an off-period.

[0022] (5) Preferably, the output signal according to the user's exercise includes a signal that causes an output during an ON period of the pulse signal. This allows an output to be generated when the load of the user's exercise exceeds a threshold. Therefore, the output allows the user to know that the load of the user's exercise exceeds the threshold.

[0023] (6) Preferably, the measurement value obtained from the user is a measurement value based on the myoelectric signal and other signals. The other signals are, for example, a signal based on the detection result of the shape of a body part such as the user's joint angle, a signal indicating a specific rhythm, a second myoelectric signal obtained from a second myoelectric sensor different from the first myoelectric sensor that measures the myoelectric signal of a second user different from the first myoelectric sensor, etc. By using the measurement value based on the myoelectric signal and other signals, it is possible to perform output based on the user's exercise, taking into account factors other than the exercise load.

[0024] (7) A biofeedback method according to an embodiment is a biofeedback method using an electromyogram signal input from an electromyogram sensor that measures the electromyogram of a user, and includes correcting a stored threshold value according to a measurement value obtained from the user, converting the value of the electromyogram signal obtained from the electromyogram sensor into a pulse signal based on a comparison with the corrected threshold value, and using the pulse signal to generate an output signal according to the user's movement. By correcting the threshold value in consideration of the user's physical state and converting the electromyogram signal into a pulse signal for use, it is possible to generate an output signal according to the user's movement with less influence of fluctuations in the value of the electromyogram signal on the calculation than by using a feature obtained from the value of the electromyogram signal.

[0025] (8) A computer program according to an embodiment is a computer program that causes a computer to execute a calculation using an electromyogram signal input from an electromyogram sensor that measures the electromyogram of a user, the calculation including correcting a stored threshold value according to a measurement value obtained from the user, converting a value of the electromyogram signal obtained from the electromyogram sensor into a pulse signal based on a comparison with the corrected threshold value, and generating an output signal according to the user's motion using the pulse signal. This allows a general-purpose computer to execute the calculations of the biofeedback system of (1) to (6).

[0026] 2. Examples of biofeedback systems, biofeedback methods, and computer programs

[0027] [First embodiment]

[0028] A biofeedback system (hereinafter abbreviated as the system) 100 according to this embodiment outputs biological information of a subject according to the exercise being performed by the subject. Here, exercise refers to a movement accompanied by muscle activity, in which the exercise load is measured by an electromyographic signal. Output refers to outputting data for output (hereinafter output data). Output data is, for example, a predetermined action such as sound, an image, vibration, or a combination of these. Sound also includes voices that express words as sound, sounds such as bell sounds without words, and combinations of these.

[0029] The output according to the exercise is, for example, an output representing an exercise load. The output representing the exercise load may be, for example, an output of a degree of deviation from a target exercise. In other words, the degree of deviation may be a degree of agreement with the target exercise. Or, it may be both a degree of agreement and a degree of deviation. In this way, the degree of deviation or the degree of agreement is notified to the user.

[0030] A target motion refers to a state of motion that is targeted for a certain motion. The state of motion refers to the amount of motion, posture during motion, speed of movement, etc. Posture during motion refers to the shape of the body during motion, such as the angle of the upper body while running or the angle of the knees while squatting.

[0031] The state of exercise, such as the amount of exercise and exercise posture, places a load on a specific muscle. For example, when the upper body is leaning forward more than the target angle while jogging, the load on the back side of the human body is higher than the load at the target angle. Therefore, the myoelectric signal at a specific position on the back side can be used to determine whether the upper body is leaning forward. Similarly, the myoelectric signal can be used to determine the speed of movement.

[0032] The system 100 is expected to be used in scenes involving exercise, such as recreation involving exercise in care facilities such as nursing homes and childcare facilities, music education through physical movements, such as eurhythmics, in music classes and special support classes, fitness classes performed in groups or remotely, and movements such as dancing and marching to a specific rhythm such as music. In such scenes, the system 100 outputs according to the degree of agreement between the subject's movement during exercise and the target movement. Alternatively, the system may output according to the degree of discrepancy, or may output according to both. The output is, for example, audio output, and as an example, a specific sound is output when the movement matches the target movement or when it deviates from it. Alternatively, different sounds are output when the movement matches and when it deviates from it.

[0033] 1, the system 100 includes an electromyogram sensor 30 that measures electromyograms, and a calculation device 10 that executes calculations using electromyogram signals input from the electromyogram sensor 30. The calculation device 10 may be, for example, a terminal device such as a smartphone, or a computer.

[0034] The myoelectric sensor 30 is attached to the body S of a subject, and measures the myoelectric potential of the body S at the site where the sensor is attached. The myoelectric sensor 30 includes a measuring unit 31 and a transmitting unit 32.

[0035] The measurement unit 31 includes electrodes (not shown). The electrodes are arranged on the outer peripheral surface of the myoelectric sensor 30 that directly contacts the measurement site of the subject when the sensor 30 is attached to the body S. This allows the myoelectric potential to be measured from the surface of the body S. The transmission unit 32 transmits a myoelectric signal including the measurement result to the calculation device 10. Preferably, the transmission unit 32 transmits the myoelectric signal by wirelessly communicating with the calculation device 10.

[0036] The system 100 further includes a device that inputs another signal to the arithmetic device 10. In the system 100 according to the first embodiment, the device that inputs another signal to the arithmetic device 10 is a device that detects the shape of a part of the body S of the subject. The shape of the part of the body S is, for example, the angle of the part. The system 100 includes an angle sensor 50 as a device that inputs a signal indicating an angle to the arithmetic device 10.

[0037] The angle sensor 50 is attached to the body S of the subject and measures the angle of the part where it is attached. The angle sensor 50 includes a measurement unit 51 and a transmission unit 52. The method of measuring the angle by the measurement unit 51 is not limited to a specific method. For example, the measurement unit 51 may measure the change in angle as a change in capacitance. The transmission unit 52 transmits a sensor signal indicating the measured angle to the calculation device 10. Preferably, the transmission unit 52 transmits the sensor signal by wirelessly communicating with the calculation device 10.

[0038] Preferably, the system 100 further includes a sensor 70 for performing measurements related to the subject's body S. The measurements related to the body S are measurements of the environment of the body S that affect the myoelectric measurement by the myoelectric sensor 30, such as humidity, temperature, and roughness of the surface of the body S. This is because the myoelectric measurement by the myoelectric sensor 30 is affected by such environmental conditions. As an example, the sensor 70 is a humidity sensor. The sensor 70 also transmits a sensor signal indicating the measured humidity to the computing device 10. Preferably, this transmission is also by wireless communication.

[0039] The arithmetic device 10 is a terminal device such as a smartphone, a server connected via a communication line such as the Internet, or the like. With reference to Fig. 2, the arithmetic device 10 is configured as a computer having a processor 11 and a memory 12. The memory 12 may be a primary storage device or a secondary storage device. The memory 12 stores a program 121 to be executed by the processor 11.

[0040] The memory 12 further includes a threshold storage unit 122 that stores one or more thresholds used in the process executed by the processor 11, which will be described later. As an example, a plurality of thresholds TH1 and TH2 are stored in the threshold storage unit 122. The memory 12 also includes a condition storage unit 123 that stores conditions used in the process executed by the processor 11, which will be described later.

[0041] The arithmetic device 10 further includes a communication device 13. The communication device 13 communicates with the myoelectric sensor 30, the angle sensor 50, and the sensor 70. Preferably, the communication device 13 wirelessly communicates with these sensors. In this case, the communication device 13 is, for example, a wireless communication module. The communication device 13 inputs sensor signals received from these sensors to the processor 11.

[0042] The computing device 10 further includes a speaker 14, which is an example of an output device. Instead of or in addition to the speaker 14, the computing device 10 may include a display. The computing device 10 may further include a camera, which may function as the sensor 70. The speaker 14 is connected to the processor 11 and outputs audio under the control of the processor 11.

[0043] The processor 11 executes the program 121 stored in the memory 12 to perform calculations using the myoelectric signal input from the myoelectric sensor 30. The calculations using the myoelectric signal are calculations for generating an audio signal, which is an output signal for causing the speaker 14 to output audio based on the myoelectric signal, and for causing the speaker 14 to output the audio signal.

[0044] The calculations executed by the processor 11 include a pulse generation process 111. The pulse generation process 111 is a process for converting the value of the myoelectric signal obtained from the myoelectric sensor 30 into a pulse signal based on a comparison with a set threshold value.

[0045] The pulse generation process 111 will be described with reference to Fig. 3. In the pulse generation process 111, a threshold stored in the threshold storage unit 122 is read out and used as the set threshold. Specifically, the pulse generation process 111 compares a waveform obtained from the value of the myoelectric signal from the myoelectric sensor 30 with the threshold stored in the threshold storage unit 122, and converts the value of the myoelectric signal obtained from the myoelectric sensor 30 into a pulse signal by setting a period in which the threshold is exceeded as an on period and a period in which the threshold is not exceeded as an off period.

[0046] The thresholds stored in the threshold storage unit 122 may be a plurality of thresholds TH1 and TH2 as shown in Fig. 2. In this case, the pulse generation process 111 generates a pulse signal using at least one of the thresholds TH1 and TH2. In the following description, as an example, the pulse generation process 111 compares a sensor value represented by a waveform SG obtained from an electromyogram signal from the electromyogram sensor 30 with each of the thresholds TH1 and TH2 stored in the threshold storage unit 122 as shown in Fig. 3.

[0047] The thresholds TH1 and TH2 are set based on the change over time in the value of the myoelectric signal when the target exercise is being performed, which is represented by waveform M. As an example, threshold TH1 is the peak value P of waveform M, as shown in FIG. 3. Threshold TH2 is set in stages based on the value of the myoelectric signal when the target exercise is being performed, and as an example, is a value twice as high as threshold TH1. By setting multiple thresholds in stages, the exercise load of the subject represented by the myoelectric signal can be detected in stages.

[0048] As another example, the pulse generation process 111 may generate a pulse signal using one of the thresholds TH1 and TH2 (for example, only the threshold TH1), or may generate a pulse signal that changes in stages using two or more of the thresholds TH1 and TH2.

[0049] In the pulse generation process 111, the processor 11 generates a pulse signal PU1 using a threshold value TH1 for the waveform M. Also, the processor 11 generates a pulse signal PU2 using a threshold value TH2. The pulse signal PU1 is a pulse waveform in which the periods t11-t12, t13-t14, t15-t16, and t17-t18 of the waveform M that exceed the threshold value TH1 are on periods, and the periods t12-t13, t14-t15, t16-t17, and t18-t19 that do not exceed the threshold value are off periods. The pulse signal PU2 is a pulse waveform in which the periods t21-t22, t23-t24, t25-t26, and t27-t28 of the waveform M that exceed the threshold value TH2 are on periods, and the periods t22-t23, t24-t25, t26-t27, and t28-t29 that do not exceed the threshold value are off periods.

[0050] The waveform SG based on the myoelectric signal input from the myoelectric sensor 30 generally has a large change in amplitude for each period, and a large fluctuation in a particular feature value such as a maximum value. In this regard, by generating a pulse signal using a threshold value, the fluctuation for each period can be suppressed. As a result, the determination accuracy in the sound identification process 112 described later can be improved compared to using the feature points of the waveform SG.

[0051] The calculations executed by the processor 11 include a sound identification process 112. The sound identification process 112 is a process for identifying corresponding output data based on the myoelectric signal converted into a pulse signal and other signals, that is, a process for identifying a sound to be output from the speaker 14.

[0052] The sound identification process 112 will be described with reference to FIG. 4. The sound identification process 112 uses conditions stored in the condition storage unit 123. The conditions are output data, that is, output conditions defined for each sound, and the condition storage unit 123 stores output conditions for each sound identification data. In the system 100 according to the first embodiment, the output condition is the shape of a part of the subject's body S. The shape of the part is, for example, an angle (joint angle). In the example of FIG. 4, the knee angle in a squat exercise is stored as the output condition for each sound name as sound identification data.

[0053] Specifically, referring to Fig. 4, the first condition C1 associates the angle α1 with sound output data of 440 Hz, which corresponds to the identification data "La". The second condition C2 associates the angle α2 with sound output data of 493.88 Hz, which corresponds to the identification data "Si". The third condition C3 associates the angle α3 with sound output data of 493.88 Hz, which corresponds to the identification data "Do". Note that the angles α1 to α3 are not limited to one angle each, and may be within an allowable range of angles.

[0054] Also, there may be only one condition. For example, a condition associated with a sound may be prepared for only the angle of the target motion or the allowable range of that angle. In this case, when the target motion is achieved, a sound is generated to notify the user, and when no sound is output, a deviation from the target motion is notified. The same applies to the other embodiments described below.

[0055] The sound identification process 112 includes comparing the sensing angle indicated in the sensor signal from the angle sensor 50 during the on period of the pulse signal generated in the pulse generation process 111 with the angles or allowable ranges of angles defined in the first condition C1 to the third condition C3. In the sound identification process 112, the processor 11 identifies a condition that corresponds to an angle equivalent to the sensing angle or an allowable range that includes the angle, and identifies the sound as outputting the output data indicated in that condition.

[0056] The calculations executed by the processor 11 include a signal generation process 113. The signal generation process 113 is a process for generating an audio signal for outputting the sound, which is the output data identified by the sound identification process 112, from the speaker 14.

[0057] The calculations executed by the processor 11 include output control processing 114. The output control processing 114 is processing for controlling output of the sound identified in the sound identification processing 112 to the speaker 14. Specifically, the output control processing 114 is processing for passing the audio signal generated in the signal generation processing 113 to the speaker 14 and instructing sound output based on the audio signal.

[0058] Preferably, the calculations executed by the processor 11 further include a correction process 115. The correction process 115 is a process for correcting the thresholds TH1 and TH2 stored in the threshold storage unit 122 in accordance with a measurement value obtained from the subject's body S by the sensor 70. The measurement value is, for example, the humidity of the surface of the body S.

[0059] As an example, the correction process 115 may be a process for obtaining corrected threshold values ​​TH1, TH2 by substituting the humidity obtained from the sensor 70 into a correction formula stored in advance, or may be a process for obtaining a correction amount based on the relationship between the humidity and the correction amount of the threshold values ​​TH1, TH2 obtained in advance by experiment or the like.

[0060] A biofeedback method in the system 100 according to the first embodiment will be described with reference to Fig. 5. The biofeedback method shown in the flowchart of Fig. 5 is realized by the processor 11 of the arithmetic device 10 executing the program 121 and carrying out each of the steps 111 to 115.

[0061] The biofeedback method in the present system 100 is carried out, for example, when instructing a subject on exercise in a care facility or the like, and is carried out by attaching sensors 30, 50, 70 to the body S of the subject as shown in Fig. 1 and setting the body S in a state in which it can communicate with a computing device 10 such as a smartphone. A program 121 is started in the computing device 10, and the processing in Fig. 5 is started when the computing device 10 is in a processing-enabled state.

[0062] 5, when starting processing according to the program 121, the processor 11 accepts an input of a myoelectric signal from the myoelectric sensor 30 (step S101). Step S101 may be to receive from the communication device 13 a myoelectric signal that has been received by the communication device 13 and transmitted by wireless communication from the myoelectric sensor 30, or to read from the memory 12 a myoelectric signal that has already been received and stored in the memory 12.

[0063] The processor 11 also receives an input of a sensor signal representing an angle from the angle sensor 50 (step S103). The sensor signal from the angle sensor 50 may also be stored in the memory 12. The sensor signal from the angle sensor 50 is associated with data indicating a temporal relationship with the myoelectric signal. As an example, a timestamp indicating the time of measurement is added to both the myoelectric signal and the sensor signal from the angle sensor 50. This allows the temporal relationship between the myoelectric signal and the sensing angle to be used in subsequent processing.

[0064] The processor 11 generates a pulse signal from the myoelectric signal input in step S101 (step S107). In other words, the generation of the pulse signal in step S107 corresponds to a first determination of whether or not to output a sound, that is, a determination of a period during which a sound is output and a period during which a sound is not output. At this time, the processor 11 reads out the thresholds TH1 and TH2 stored in the threshold storage unit 122 and uses them as shown in FIG.

[0065] Preferably, the processor 11 corrects the thresholds TH1 and TH2 used in generating the pulse signal in step S107 using a value measured from the subject's body S, such as the humidity on the surface of the body S (step S105). This improves the accuracy of the pulse signal generated in step S107. Note that a similar correction may be performed in the biofeedback methods according to the second and third embodiments described below.

[0066] When the pulse signal is generated in step S107, the processor 11 identifies the sound using the sensing angle indicated in the sensor signal input from the angle sensor 50 in step S103 (step S109). In other words, the identification of the sound in step S109 corresponds to a second judgment for judging the sound to be output, that is, determining the sound to be output. In step S109, the processor 11 reads out the conditions stored in the condition storage unit 123, and identifies the sound for which the sensing angle measured during the on-period of the pulse signal generated in step S107 is within the allowable range.

[0067] When the sound is identified in step S109, the processor 11 generates an audio signal for outputting the identified sound (step S111). The processor 11 passes the audio signal to the speaker 14 to instruct it to output, thereby causing the speaker 14 to output the sound (step S113).

[0068] In the present system 100, the biofeedback method described above converts the myoelectric signal obtained from the subject's body S into a pulse signal and uses the pulse signal for calculation, which can reduce the effect of fluctuations in the myoelectric signal value on the calculation, compared to using features obtained from the myoelectric signal value. Therefore, it is possible to output a sound according to the user's movement based on the myoelectric signal value.

[0069] Furthermore, with the above biofeedback method, the system 100 can output a sound based on the detection result of the shape of the body S in addition to the myoelectric signal. Even if a high-stress exercise is performed, the exercise may differ from the body shape of the target exercise. For example, in the example of FIG. 4, the knee angle of the target exercise during squatting is angle α2 represented by condition C2. In this case, when the exercise load is equal to or greater than the load of the target exercise and the knee angle is angle α2, the sound "shi" is output from the speaker 14. Therefore, the user can know that the squatting of the subject is close to the target exercise by hearing the sound "shi".

[0070] If the knee angle is larger (condition C1) or smaller (condition C2) than angle α2, the shape of the target movement deviates even if the load of the movement is equal to or greater than that of the target movement. In this case, the user hears the sound of "ra" or "do" other than "si", and can tell from the sound that the subject's squat deviates from the target movement.

[0071] [Second embodiment]

[0072] Another example of the other signal may be a signal indicating a specific rhythm. In the system 100 according to the second embodiment, the arithmetic device 10 includes or is connected to, for example, an oscillator (not shown) and receives an input of a signal indicating a specific rhythm. In this case, in the sound identification process 112, the processor 11 of the arithmetic device 10 recognizes the sound to be output based on the timing of the movement obtained from the myoelectric signal and the input specific rhythm during the on-period of the pulse signal.

[0073] 6, in the sound identification process 112, the processor 11 determines, as an example, the timings t11, t13, t15, and t17 of the start of the on-period of the pulse signal generated in the pulse generation process 111 as the timings of the movement obtained from the pulse signal based on the myoelectric signal. The processor 11 obtains time differences s11 to s14 by comparing the obtained timings with timings T1, T2, T3, and T4 indicated in the input specific rhythm.

[0074] In this case, the conditions stored in the condition storage unit 123 prescribe, for each sound, a time difference from the timing indicated by a specific rhythm or a tolerable range of the time difference. The processor 11 identifies the sound that is prescribes the conditions corresponding to the time differences s11 to s14.

[0075] A biofeedback method in the system 100 according to the second embodiment will be described with reference to Fig. 7. When processing according to the program 121 is started with reference to Fig. 7, the processor 11 accepts an input of a myoelectric signal from the myoelectric sensor 30 (step S201). The processor 11 also accepts an input of a rhythmic signal (step S203). In step S203, the processor 11 accepts an input of a rhythmic signal from, for example, an oscillator (not shown) mounted on the arithmetic device 10. The processor 11 generates a pulse signal from the myoelectric signal input in step S201 (step S207).

[0076] In the biofeedback method according to the second embodiment, the processor 11 identifies the timing of the subject's exercise from a specific timing of the on-period of the pulse signal generated in step S207. As an example, the timing of the exercise is set to start times t11, t13, t15, and t17 of the on-period of the pulse signal.

[0077] When the pulse signal is generated in step S207, the processor 11 identifies sounds based on the rhythm indicated by the rhythm signal input in step S203 (step S209). In step S209, the processor 11 compares the timing of the movement obtained from the pulse signal with the rhythm indicated by the rhythm signal to obtain time differences s11 to s14. The processor 11 then identifies sounds for which an allowable range including the obtained time differences s11 to s14 is specified, among the allowable ranges of time differences specified for each sound of the conditions stored in the condition storage unit 123.

[0078] When the sound is identified in step S209, the processor 11 generates an audio signal for outputting the identified sound (step S211). The processor 11 passes the audio signal to the speaker 14 to instruct it to output, thereby causing the speaker 14 to output the sound (step S213).

[0079] In the system 100 according to the second embodiment, the biofeedback method described above can output a sound in accordance with the degree of agreement with a specific rhythm in addition to the myoelectric signal. In this case, the exercise includes, for example, dance, which requires movements in time with music.

[0080] In such exercise, even if a high-stress exercise is performed, if the exercise does not match the rhythm of the target exercise, it deviates from the target exercise. In this regard, in the present system 100, a sound is output according to the degree of deviation of the timing of the exercise of the subject from the rhythm. That is, if the rhythm matches the rhythm obtained from the music, a sound indicating this is output from the speaker 14. Conversely, if the rhythm does not match, a sound indicating this is output from the speaker 14. Therefore, the user can know the degree of deviation of the timing of the exercise of the subject from the rhythm by the sound.

[0081] [Third embodiment]

[0082] As another example of the other signal, it may be a myoelectric signal from another myoelectric sensor. In the system 100 according to the third embodiment, the arithmetic device 10 receives an input of a myoelectric signal from another myoelectric sensor (not shown) different from the myoelectric sensor 30.

[0083] In this case, in the sound identification process 112, the processor 11 of the arithmetic device 10 identifies the sound to be output based on the timing of movement obtained from the myoelectric signal during the on period of the pulse signal and the timing of movement obtained from the input myoelectric signals from other myoelectric sensors.

[0084] 8, in pulse generation processing 111, the processor 11 generates a pulse signal PU1 from a myoelectric signal SG1 obtained from the myoelectric sensor 30, and also generates a pulse signal PU3 in a similar manner from a myoelectric signal SG2 obtained from another myoelectric sensor. The threshold value TH1 used here may be the same threshold value, or may be a different threshold value depending on the subject or the part to which the myoelectric sensor is attached.

[0085] In the sound identification process 112, the processor 11, for example, sets the timings t11, t13, t15, and t17 of the start of the on-period of the pulse signal PU1 generated in the pulse generation process 111 as the timings of the movement obtained from the pulse signal PU1. Similarly, the timings t31, t33, t35, and t37 of the start of the on-period of the pulse signal PU3 as the timings of the movement obtained from the pulse signal PU3. Then, the processor 11 compares the timings t11, t13, t15, and t17 with the timings t31, t33, t35, and t37, respectively. The processor 11 obtains the time differences s21 to s24 by comparing the timings t11, t13, t15, and t17 with the timings t31, t33, t35, and t37, respectively.

[0086] In this case, the conditions stored in the condition storage unit 123 define, for each sound, the time difference between the timing of the movement obtained from each of the two myoelectric signals, or the allowable range of the time difference. The processor 11 identifies the sound defined by the conditions corresponding to the time differences s21 to s24.

[0087] A biofeedback method in the system 100 according to the third embodiment will be described with reference to Fig. 9. When processing according to the program 121 is started with reference to Fig. 9, the processor 11 receives an input of a myoelectric signal from the myoelectric sensor 30 (step S301) and also receives an input of myoelectric signals from other myoelectric sensors (step S303). The processor 11 generates pulse signals PU1 and PU3 from the myoelectric signals received in steps S301 and S303, respectively (step S307).

[0088] When the pulse signals PU1 and PU3 are generated in step S307, the processor 11 identifies a sound based on both the pulse signals PU1 and PU3 generated in step S307 (step S309).

[0089] In the biofeedback method according to the third embodiment, the processor 11 identifies the timing of movement from specific timings of the on-periods of the pulse signals PU1 and PU3 generated in step S307. As an example, the timings of movement obtained from the pulse signal PU1 are t11, t13, t15, and t17, and the timings of movement obtained from the pulse signal PU3 are t31, t33, t35, and t37.

[0090] In step S309, processor 11 obtains time differences s21 to s24 by comparing timings t11, t13, t15, and t17 with timings t31, t33, t35, and t37. Processor 11 then identifies sounds for which an allowable range of time differences that includes the obtained time differences s21 to s24 is defined, among the allowable ranges of time differences defined for each sound of the conditions stored in condition storage unit 123.

[0091] When the sound is identified in step S309, the processor 11 generates an audio signal for outputting the recognized sound (step S311). The processor 11 passes the audio signal to the speaker 14 to instruct it to output, thereby causing the speaker 14 to output the sound (step S313).

[0092] According to the above biofeedback method, the system 100 according to the third embodiment can output a sound according to the exercise load and the degree of agreement between the timing of exercise obtained from a plurality of myoelectric signals.

[0093] In one example of a method of use, a myoelectric sensor is attached to each of the bodies of a plurality of subjects, and the computing device 10 performs a calculation using each of the myoelectric signals. In this case, the system 100 outputs a sound according to the degree of agreement between the timing of the exercise by the subject and the timing of the exercise by other users. As the exercise in this case, a timed exercise such as dancing or marching is assumed.

[0094] Another example of the usage method is to attach myoelectric sensors to multiple locations on the body S of the same subject, and perform calculations using the respective myoelectric signals in the calculation device 10. In this case, in the system 100, during exercise by the same subject, sounds are output according to the degree of agreement between the timing of the exercise of a target part and the timing of the exercise of other parts. The exercise in this case is an exercise that requires the movement of multiple parts of the body at related timings, such as exercise form or dance choreography.

[0095] In such exercise, even if a high-load exercise is performed, if the timing of the exercise does not match the timing of the exercise of other users or the timing of the exercise of other parts as the target exercise, the exercise deviates from the target exercise. In this regard, in the present system 100, a sound is output according to the degree of deviation of the exercise of the subject from the target exercise. That is, if the timing of the exercise matches the timing of the exercise of other users or other parts, a sound indicating this is output from the speaker 14. Conversely, if the timing of the exercise does not match the timing of the exercise of other users or other parts, a sound indicating this is output from the speaker 14. Therefore, the user can know the degree of deviation of the timing of the exercise of the subject from the timing of the exercise of other users or other parts by the sound.

[0096] <3. Notes> Although the processing of the disclosed arithmetic device 10 has been described above as being realized by the processor 11 executing the program 121, at least a portion of it may be realized by circuit elements or other hardware.

[0097] The program 121 for implementing the processing in the arithmetic device 10 may be provided as a program product recorded on a computer-readable non-transitory recording medium. Alternatively, the program 121 may be provided by downloading via a network.

[0098] A program 121 for implementing processing in the arithmetic device 10 has program code for processing in the arithmetic device 10. The program 121 is read and executed by the processor 11 connected to the memory 12 in which the program 121 is stored.

[0099] The program 121 that realizes the processing in the arithmetic device 10 may be provided as an application program, or a part or the whole of it may be included in the operating system (OS) of the computer.

[0100] The present invention is not limited to the above-described embodiment, and various modifications are possible. [Explanation of symbols]

[0101] 10: Arithmetic device 11: Processor 12: Memory 13: Communication equipment 14: Speaker 30:Electromyography sensor 31: Measuring part 32: Transmitter 50: Angle sensor 51: Measuring part 52: Transmitter 70: Sensor 100: System 111: Pulse generation processing 112: Sound Identification Processing 113: Signal generation processing 114: Output control processing 115: Correction processing 121: Program 122: Threshold memory unit 123: Condition memory section C1: First condition C2: Second condition C3: The third condition M: Waveform P: Peak value PU1: Pulse signal PU2: Pulse signal PU3: Pulse signal S: body SG: Waveform SG1: EMG signal SG2: EMG signal T1: Timing T2: Timing T3: Timing T4: Timing TH1: Threshold TH2: Threshold s11: Time difference s12: Time difference s13: Time difference s14: Time difference s21: Time difference s22: Time difference s23: Time difference s24:Time difference t11: Timing t12: Timing t13: Timing t14: Timing t15: Timing t16: Timing t17: Timing t21: Timing t22: Timing t23: Timing t24: Timing t25: Timing t26: Timing t27: Timing t31: Timing t33: Timing t34: Timing t35: Timing t36: Timing t37: Timing α1 :Angle α2 :Angle α3 :Angle

Claims

1. A biofeedback system including a computing device that performs a calculation using an electromyogram signal input from an electromyogram sensor that measures an electromyogram of a user, The computing device includes: correcting a stored threshold value according to a measurement value obtained from the user separately from the myoelectric signal; converting a value of a myoelectric signal obtained from the myoelectric sensor into a pulse signal having a pulse width corresponding to a period during which the myoelectric signal exceeds the threshold based on a comparison with the corrected threshold; using the pulse signal to generate an output signal responsive to movement of the user; The measurement value is a measurement value indicating the environment of the skin surface of the user. Biofeedback system.

2. Correcting the stored threshold value includes correcting the stored threshold value using the measured value obtained from the user during measurement by the myoelectric sensor.

2. The biofeedback system of claim 1.

3. The threshold value is a predetermined value obtained from a time change in the value of the myoelectric signal when a prescribed exercise is performed.

3. A biofeedback system according to claim 1 or 2.

4. The conversion to the pulse signal includes setting a period during which the waveform, which is indicated by a change over time in the value of the myoelectric signal, exceeds the threshold to a constant value. A biofeedback system according to any one of claims 1 to 3.

5. The output signal according to the user's motion includes a signal that causes an output to be performed during an ON period of the pulse signal. A biofeedback system according to any one of claims 1 to 4.

6. A method for performing biofeedback using an electromyogram signal input from an electromyogram sensor that measures an electromyogram of a user, comprising: correcting a stored threshold value according to a measurement value obtained from the user separately from the myoelectric signal; converting a value of a myoelectric signal obtained from the myoelectric sensor into a pulse signal having a pulse width corresponding to a period during which the myoelectric signal exceeds the threshold based on a comparison with the corrected threshold; using the pulse signal to generate an output signal responsive to movement of the user; The measurement value is a measurement value indicating the environment of the skin surface of the user. Biofeedback methods.

7. A computer program for causing a computer to execute a calculation using an electromyogram signal input from an electromyogram sensor that measures an electromyogram of a user, correcting a stored threshold value according to a measurement value obtained from the user separately from the myoelectric signal; converting a value of a myoelectric signal obtained from the myoelectric sensor into a pulse signal having a pulse width corresponding to a period during which the myoelectric signal exceeds the threshold based on a comparison with the corrected threshold; using the pulse signal to generate an output signal responsive to movement of the user; The measurement value is a measurement value indicating the environment of the skin surface of the user. Computer program.

Images