JP2021137417A - Computer program, muscle function parameter calculation device, muscle function parameter calculation system, and muscle function parameter calculation method - Google Patents

Abstract

To provide a muscle function parameter calculation device for calculating muscle function parameters indicating a muscle function of a user.SOLUTION: A muscle function parameter calculation device 2 includes: an acquisition unit for acquiring sensor data from a wearable device 1 that is attached to the body of a user A; and a calculation unit for calculating muscle function parameters indicating at least the level of the muscle function, on the basis of the acquired sensor data.SELECTED DRAWING: Figure 1

Description

The present invention relates to a computer program, a muscle function parameter calculation device, a muscle function parameter calculation system, and a muscle function parameter calculation method.

A method of measuring biological information during exercise of a user and estimating the exercise ability of the user based on the measurement result is disclosed (for example, Patent Document 1).

Japanese Unexamined Patent Publication No. 2016-32525

An object of the present invention is to provide a computer program for calculating muscle function parameters indicating a user's muscle function, a muscle function parameter calculation device, a muscle function parameter calculation system, and a muscle function parameter calculation method.

The computer program according to this aspect acquires sensor data from a wearable device worn on the user's body, and calculates at least muscle function parameters indicating the level of the user's muscle function based on the acquired sensor data. Let the computer do it.

The muscle function parameter calculation device according to this embodiment has an acquisition unit that acquires sensor data from a wearable device attached to the user's body, and at least the user's muscle function based on the sensor data acquired by the acquisition unit. It is provided with a calculation unit that calculates muscle function parameters indicating high and low.

The muscle function parameter calculation system according to this embodiment includes a wearable device having an acceleration sensor, an acquisition unit that acquires sensor data from the wearable device attached to the user's body, and the sensor data acquired by the acquisition unit. Based on this, it is provided with a muscle function parameter calculation device having at least a calculation unit for calculating muscle function parameters indicating the level of the user's muscle function.

In the muscle function parameter calculation method according to this aspect, sensor data is acquired from a wearable device worn on the user's body, and based on the acquired sensor data, at least muscle function parameters indicating the level of the user's muscle function are calculated. do.

According to the above, it is possible to provide a computer program for calculating muscle function parameters indicating a user's muscle function, a muscle function parameter calculation device, a muscle function parameter calculation system, and a muscle function parameter calculation method.

It is a schematic diagram explaining the structural example of the muscle function parameter calculation system which concerns on Embodiment 1. FIG. It is a block diagram which shows the configuration example of the wearable device which concerns on Embodiment 1. FIG. It is a block diagram which shows the structural example of the muscle function parameter calculation apparatus which concerns on Embodiment 1. It is a flowchart which shows the processing procedure of a control part. It is a flowchart which shows the processing procedure of a control part. It is a graph which shows the acceleration waveform of an impact operation. It is a graph which shows the acceleration waveform of an impact operation. It is explanatory drawing explaining the method of calculating the impact strength, impact property and impact force based on the acceleration data at the time of impact operation. It is a graph which shows the acceleration waveform of the stretch stop. It is a graph which shows the acceleration waveform of the stretch stop. It is explanatory drawing explaining the method of calculating the muscle strength strength, instantaneousness and muscle strength based on the acceleration data at the time of stopping stretching. It is explanatory drawing explaining the method of calculating the agility intensity, agility and agility based on the acceleration data at the time of stopping stretching. It is a graph which shows the acceleration waveform of the expansion and contraction operation. It is a graph which shows the acceleration waveform of the expansion and contraction operation. It is explanatory drawing explaining the method of calculating muscle strength strength, instantaneousness and muscle strength, and agility strength, agility and agility based on acceleration data at the time of expansion and contraction operation. It is a schematic diagram which shows the display example of a muscle function parameter. It is a schematic diagram which shows the display example of the acceleration waveform. It is a functional block diagram of the muscle function parameter calculation apparatus which concerns on Embodiment 2. FIG. It is a functional block diagram of the muscle function parameter calculation apparatus which concerns on Embodiment 3.

Specific examples of the computer program, the muscle function parameter calculation device, the muscle function parameter calculation system, and the muscle function parameter calculation method according to the embodiment of the present invention will be described below with reference to the drawings. It should be noted that the present invention is not limited to these examples, and is indicated by the scope of claims, and is intended to include all modifications within the meaning and scope equivalent to the scope of claims.

Hereinafter, the present invention will be specifically described with reference to the drawings showing the embodiments thereof.
(Embodiment 1)
<Muscle function parameter calculation system>
FIG. 1 is a schematic diagram illustrating a configuration example of a muscle function parameter calculation system according to the first embodiment. The muscle function parameter calculation system according to the first embodiment includes a wearable device 1 and a muscle function parameter calculation device 2. The muscle function parameter calculation device 2 can communicate with the server 3 via the communication network N.

FIG. 2 is a block diagram showing a configuration example of the wearable device 1 according to the first embodiment. The wearable device 1 includes an arithmetic processor 10, an acceleration sensor 11, and a transmission unit 12.

The acceleration sensor 11 detects the acceleration according to the movement of the user A wearing the wearable device 1.
In the first embodiment, the wearable device 1 will be described as being worn on the arm. When the wearable device 1 is attached to the arm of the user A, the acceleration sensor 11 detects the acceleration according to the movement of the arm. It is desirable that the acceleration measurement cycle by the acceleration sensor 11 is, for example, 50 msec or less.

The arithmetic processor 10 processes the acceleration data detected by the acceleration sensor 11. The data processing is a data conversion process. The muscle function parameter calculation device 2 calculates the muscle function parameter. That is, the muscle function parameters that the user A can recognize as useful information are calculated on the muscle function parameter calculation device 2 side. The arithmetic processor 10 executes preprocessing for processing acceleration data so that the muscle function parameter calculation device 2 can efficiently process the muscle function parameters.

Further, the arithmetic processor 10 does not output all the acceleration data to the transmission unit 12, but outputs acceleration data of a predetermined threshold value or more to the transmission unit 12.

The transmission unit 12 includes a memory 12a and a wireless communication unit 12b. The memory 12a stores or stores information indicating the acceleration detected by the acceleration sensor 11 and the measurement time point.

The wireless communication unit 12b wirelessly transmits sensor data including information indicating the acceleration and the measurement time point stored in the memory 12a to the muscle function parameter calculation device 2. After the acceleration monitoring is completed, the wireless communication unit 12b may wirelessly transmit the sensor data stored in the memory 12a. The wireless communication unit 12b wirelessly transmits sensor data, for example, by wireless communication compliant with IEEE802.5.1, that is, Bluetooth (registered trademark).

In the above description, an example in which the acceleration is stored in the memory 12a and wirelessly transmitted has been described, but the acceleration data may be configured to be wirelessly transmitted in real time.

<Muscle function parameter calculation device 2>
FIG. 3 is a block diagram showing a configuration example of the muscle function parameter calculation device 2 according to the first embodiment. The muscle function parameter calculation device 2 is, for example, a portable device such as a smartphone, a mobile phone, a tablet terminal, or a PDA (Personal Digital Assistant) owned by the user A. Hereinafter, it is assumed that the muscle function parameter calculation device 2 is a wireless communication device having a telephone function. The muscle function parameter calculation device 2 is a computer including a control unit 20 that controls the operation of each component of the own machine.

The control unit 20 is a microcomputer having a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), a clock unit, an input / output interface, and the like. A storage unit 21, a wireless communication unit 22, a short-range communication unit 23, a vibrator 24, a display unit 25, an operation unit 26, a speaker 27, a microphone 28, and the like are connected to the input / output interface of the control unit 20. The ROM stores the initial operation program required for the initial operation of the computer. The RAM is a memory such as a DRAM (Dynamic RAM) or a SRAM (Static RAM), and is a computer program 21a described later or a control unit 20 read from the storage unit 21 when executing the arithmetic processing of the control unit 20. Temporarily stores various data generated by arithmetic processing. By executing the computer program 21a, the CPU controls the operation of each component and executes a process of calculating the muscle function parameters of the user A.

The storage unit 21 is a non-volatile memory such as an EEPROM (Electrically Erasable Programmable ROM) or a flash memory. The storage unit 21 stores the computer program 21a required for the process of calculating the muscle function parameters based on the sensor data wirelessly transmitted from the wearable device 1.

The computer program 21a according to the present embodiment may be recorded on a recording medium so that it can be read by a computer. The storage unit 21 stores the computer program 21a read from the recording medium by a reading device (not shown). The recording medium is a semiconductor memory such as a flash memory. The recording medium may be an optical disk such as a CD (Compact Disc) -ROM, a DVD (Digital Versatile Disc) -ROM, or a BD (Blu-ray (registered trademark) Disc). Further, the recording medium may be a flexible disk, a magnetic disk such as a hard disk, a magnetic optical disk, or the like. Furthermore, the computer program 21a according to the present embodiment may be downloaded from an external computer connected to the communication network N, for example, a server 3, and stored in the storage unit 21.

The wireless communication unit 22 is a communication device for transmitting and receiving various information to and from another wireless communication device, a server 3 connected to the communication network N, and the like via a base station. The wireless communication unit 22 transmits / receives call voice and various data using, for example, a 3G (Generation) line, a 4G line, an LTE (Long Term Evolution) line, an LPWA (Low Power Wide Area) line, or the like.

The short-range communication unit 23 is a communication device that performs wireless communication with the wearable device 1. The short-range communication unit 23 performs wireless communication conforming to, for example, a predetermined standard, IEEE802.5.1.

The vibrator 24 includes a vibration motor or the like, and vibrates the muscle function parameter calculation device 2 according to the control of the control unit 20. For example, the control unit 20 intermittently operates the vibrator 24 when a telephone call arrives, an alarm is set, or the like.

The display unit 25 is a liquid crystal panel, an organic EL display, electronic paper, a plasma display, or the like. The display unit 25 displays various information according to the video data given by the control unit 20.

The operation unit 26 is, for example, a touch sensor, a mechanical operation button, or the like provided on the surface or inside of the display unit 25. The touch sensor receives the operation of the user A by detecting that the finger of the user A touches the display unit 25, the position where the finger touches the display unit 25, and the like, and gives the received operation information to the control unit 20. That is, the control unit 20 can accept the operation of the user A by the operation unit 26.

The speaker 27 converts the audio data given by the control unit 20 into sound waves and outputs the sound waves.
The microphone 28 converts sound waves into audio data, and gives the converted audio data to the control unit 20.

<Muscle function parameter calculation processing method>
4 and 5 are flowcharts showing the processing procedure of the control unit 20. The following processing will be described assuming that the wearable device 1 is worn on the body of the user A, specifically, the arm, and the wearable device 1 periodically wirelessly transmits sensor data.

The control unit 20 receives the sensor data transmitted from the wearable device 1 at the short-range communication unit 23 (step S11). The sensor data includes, for example, acceleration data according to the movement of the arm of the user A and measurement time point information.
The control unit 20 that executes the process of step S11 functions as an acquisition unit that acquires sensor data from the wearable device 1 mounted on the body of the user A.

Next, the control unit 20 removes noise from the sensor data (step S12). User A may take a break at his / her own will even during exercise. Acceleration sensor data measured during a break is noise that cannot be used as data for calculating user A's muscle strength, instantaneous power, muscle strength, impact strength, impact resistance, impact force, agility strength, agility, and instantaneous force. Is. The control unit 20 removes such noise. For example, when detecting the muscle strength, instantaneousness, muscle strength, impact strength, impact strength, impact force, agility strength, agility, agility, etc. of the arm of the user A, the control unit 20 is predetermined according to the detection target. Acceleration data below a predetermined value, non-pulse waveforms, etc. are removed as noise.

Next, the control unit 20 determines the type of operation of the user A based on the acquired sensor data (step S13). Types of movements of user A include impact movements, extension stop operations (stop after extension of limbs), expansion and contraction operations (combination of limb extension operations and pulling operations), and the like. It is thought that the movements of human limbs can be almost expressed by these three movements.
In addition to the above three operations, the control unit 20 may be configured to determine an independent pulling operation, that is, a pulling operation starting from a stationary operation.

The impact operation refers to an operation in which the user A moves a limb or the like, and the limb hits an arbitrary object to suddenly decelerate. Examples of the impact motion include a foot swelling motion from a raised leg to the floor, a sandback striking motion in boxing, a punching motion accompanied by a blow to the human body, an attacking action such as volleyball, and the like.

The extension stop operation is an operation in which the user A extends and stops the limbs. Examples of the stretch-stopping motion include a sizing motion in martial arts.

The expansion / contraction operation refers to an operation in which the user A extends and pulls a limb or the like. The expansion / contraction motion is composed of two types of motions, a continuous stretching motion and a pulling motion. Since the stretching operation and the pulling operation are continuously performed in the stretching operation, the pulse shape obtained from the acceleration data of the stretching operation is slightly different from the pulse shape during the stretching operation obtained from the acceleration data of the stretching stop operation. Therefore, the expansion / contraction operation is considered separately from the extension / stop operation.
Examples of the expansion and contraction movement include shadow boxing. The stretching operation and the pulling operation are operations in which the acceleration directions (acceleration signs) are opposite. Although the strength and direction of movement are different, most of the movements of the limbs are this expansion and contraction movement. Due to human motion characteristics, the stretching motion tends to have a larger acceleration when decelerating than when accelerating in the stretching direction. Similarly, in the pulling operation, the acceleration when the speed in the pulling direction is decelerated tends to be larger than the acceleration when accelerating in the pulling direction.

The control unit 20 that has completed the process of step S13 calculates various muscle function parameters according to the type of operation of the user A by the processes of steps S14 to S20.
Hereinafter, for convenience of explanation, a muscle function calculation process (step S14) when the user A performs an impact operation, a muscle function calculation process (step S15-step S16) when the user A performs an extension stop operation, and an expansion / contraction operation. The muscle function calculation process (step S17-step S18) when the stretching motion is performed and the muscle function calculation process (step S19-step S20) when the stretching motion is performed will be described in order. In the process, the control unit 20 may selectively execute any one of steps S14 and S20 according to the determination result of step S13, that is, the type of operation of the user A.

When it is determined that the type of motion is impact motion based on the acceleration data, the control unit 20 determines the impact strength, impact property, and impact force as muscle function parameters of the user A based on the acceleration data of the valley portion of the pulse. Is calculated (step S14).

6 and 7 are graphs showing the acceleration waveform of the impact operation. The horizontal axis shows time and the vertical axis shows acceleration. 6 and 7 are obtained, for example, when user A performs sandback striking. 6 and 7 show changes in acceleration during movement until the fist is moved, the fist collides with the sandbag, and the fist stops. The acceleration on the vertical axis indicates the acceleration in the direction in which the fist moves.
As shown in FIG. 7, the mountain-shaped pulse that first appears at the start of the motion indicates that the fist is accelerating toward the sandback. The acceleration of the fist is increasing in the front part of the mountain part. At the rear part of the mountain part, the acceleration decreases because the fist movable area limit is approached. During this time, the fist continues to accelerate. The part where the acceleration becomes zero indicates the state where the fist collides with the sandbag. At this time, the speed of the fist is maximized. The valley-shaped pulse indicates that the fist collides with the sandback and the fist is decelerating rapidly. Since the fist that collides with the sandback is held by the sandback, the sudden deceleration of acceleration is alleviated, and then the acceleration becomes zero.

FIG. 8 is an explanatory diagram illustrating a method of calculating impact strength, impact property, and impact force based on acceleration data during impact operation. Similar to FIGS. 6 and 7, the horizontal axis represents time and the vertical axis represents acceleration. In FIG. 8, the α point indicates the time point (zero cross point) when the acceleration changes from positive to negative. That is, the α point indicates the time when the limbs of the user A collide with the object and start sudden deceleration. In FIG. 8, the β point indicates the time point at which the acceleration is finished after the sudden deceleration.

The impact strength, impact property and impact force are represented by, for example, the following formulas (1), (2) and (3).
Impact strength = T1 x Σ (acceleration of pulse valley) ... (1)
Impact resistance = T2 × (Tβ-Tα)… (2)
Impact force = T3 x Σ (acceleration of pulse valley) / (Tβ-Tα) ... (3)
However, T1, T2 and T3 are constants for adjusting the values of impact strength, impact property and impact force so as to be numerical values of 2 to 4 digits.

Σ (acceleration of the pulse valley portion) in the above equation is a cumulative value of acceleration from the time point Tα (point α) to the time point Tβ (point β). The cumulative value and the impact intensity are values corresponding to the area of the valley portion of the pulse in FIG. The impact property is a parameter related to the time corresponding to the pulse width of the valley pulse, and it is considered that the smaller the value, the higher the degree of impact.
The impact force is the value obtained by dividing the impact strength by the impact property (value). It can be said that the larger the value of the impact force, the higher the muscle function of the user A. That is, it can be said that the higher the impact strength, the higher the muscle function is maintained when the movement is completed in a short time.

The control unit 20 may round off or round off the values calculated by the above equations (1) to (3) after the decimal point. Further, the control unit 20 may calculate the impact intensity by multiplying the peak value (minimum value) of the pulse valley portion and the pulse width.

When it is determined that the type of operation is non-stretching based on the acceleration data, the control unit 20 uses the acceleration data of the mountain-shaped pulse (pulse peak portion) that first appears in the non-stretching operation to determine that the user A As muscle function parameters, muscle strength, instantaneousness, and muscle strength are calculated (step S15). Further, the control unit 20 calculates the agility intensity, agility, and agility based on the acceleration data of the valley-shaped pulse (pulse valley portion) that appears next to the mountain-shaped pulse in the extension stop operation (step). S16).

9 and 10 are graphs showing the acceleration waveforms of the stretch stop. The horizontal axis shows time and the vertical axis shows acceleration. 9 and 10 are obtained when the user A performs, for example, a dimension stop operation. 9 and 10 show the change in acceleration during operation until the fist is extended and stopped at a certain position. The acceleration on the vertical axis generally indicates the acceleration in the direction in which the fist moves.
As shown in FIG. 10, the peak portion of the pulse that first appears at the start of the operation indicates that the arm of the user A is extended and the fist is accelerating forward. The speed of the fist is increasing in the front part of the mountain part. At the rear part of the mountain part, the acceleration decreases because the fist movable area limit is approached. The valley part of the pulse that appears next to the mountain part indicates that the fist is suddenly decelerating due to the action of stopping the fist. The small mountain-shaped pulse that appears next to the valley indicates that the fist is coasting forward even if User A intends to stop the fist.

FIG. 11 is an explanatory diagram illustrating a method of calculating muscle strength strength, instantaneousness, and muscle strength based on acceleration data at the time of stopping stretching. Similar to FIGS. 9 and 10, the horizontal axis represents time and the vertical axis represents acceleration. In FIG. 11, the point A is the start point of the stretching motion in which the limbs of the user A, for example, the fist, start accelerating. Point B indicates the time when the acceleration changes from positive to negative (zero cross point). The speed of the fist is maximized at point B.

The muscle strength strength, instantaneousness, and muscle strength based on the acceleration data of the peak portion of the pulse that first appears in the stretch-stopping motion are represented by the following equations (4), (5), and (6).
Muscle strength = T4 x Σ (acceleration of pulse peak) ... (4)
Instantaneousness = T5 × (TB-TA) ... (5)
Muscle strength = T6 x Σ (acceleration of pulse peak) / (TB-TA) ... (6)
However, T4, T5 and T6 are constants for adjusting the values of muscle strength, instantaneousness and muscle strength so as to be numerical values of 2 to 4 digits.

Σ (acceleration of the pulse peak portion) in the above equation is a cumulative value of acceleration from the time point TA (point A) to the time point TB (point B). The cumulative value and muscle strength are values corresponding to the area of the peak portion of the pulse that first appears in FIG. Instantaneousness is a parameter related to time, and it is considered that the smaller the value, the higher the degree of instantaneous force.
Muscle strength is the strength of muscle strength divided by the instantaneousness (value). It can be said that the larger the value of muscle strength, the higher the muscle function of user A. That is, it can be said that a person who has a large muscle strength and completes the movement in a short time retains a high muscle function.

The control unit 20 may round down or round off the values calculated by the above equations (4) to (6) after the decimal point. Further, the control unit 20 may calculate the muscle strength by multiplying the peak value (maximum value) of the pulse peak portion and the pulse width.

FIG. 12 is an explanatory diagram illustrating a method of calculating agility intensity, agility, and agility based on acceleration data at the time of stopping stretching. This stretching operation also has a pulse valley portion similar to the impact operation. As in FIG. 11, the horizontal axis represents time and the vertical axis represents acceleration. In FIG. 12, point B indicates the time point (zero cross point) when the acceleration changes from positive to negative after the stretching operation is performed. Point C is the time point when the acceleration changes from negative to positive (zero cross point), and is the end point of the stop operation, that is, the end point of the extension stop operation.

The agility intensity, agility and agility based on the acceleration data of the valley portion of the pulse in the stretch stop operation are represented by the following equations (7), (8) and (9).
Agility strength = T7 x Σ (acceleration of pulse valley) ... (7)
Agility = T8 × (TC-TB) ... (8)
Agility = T9 x Σ (acceleration of pulse valley) / (TC-TB) ... (9)
However, T7, T8 and T9 are constants for adjusting the agility intensity, agility and agility values to be 2 to 4 digit values.

Σ (acceleration of the pulse valley portion) in the above equation is a cumulative value of acceleration from the time point TB (point B) to the time point TC (point C). The cumulative value and the agility intensity are values corresponding to the area of the valley portion of the pulse in FIG. Agility is a parameter related to time, and it is considered that the smaller the value, the higher the degree of instantaneous force.
Agility is the value obtained by dividing the agility intensity by the agility (value). As for the agility, it can be said that the larger the value, the higher the muscle function of the user A. In other words, it can be said that those who have great agility and strength and complete the movement in a short time retain higher muscle function.

The control unit 20 may round down or round off the values calculated by the above equations (7) to (9) after the decimal point. Further, the control unit 20 may calculate the agility intensity by multiplying the peak value (minimum value) of the pulse valley portion and the pulse width.

When it is determined that the type of motion is expansion / contraction motion based on the acceleration data, the control unit 20 sets the muscle function parameter of the user A based on the acceleration data of the peak portion of the pulse in the extension motion constituting the expansion / contraction motion. , Muscle strength strength, instantaneousness and muscle strength are calculated (step S17). Further, the control unit 20 calculates the agility intensity, agility, and agility based on the acceleration data of the valley portion of the pulse in the stretching motion constituting the stretching motion (step S18).
Next, the control unit 20 calculates the muscle strength, the instantaneousness, and the muscle strength as the muscle function parameters of the user A based on the acceleration data of the valley portion of the pulse in the pulling motion constituting the expansion / contraction motion (step S19). Further, the control unit 20 calculates the agility intensity, agility and agility based on the acceleration data of the peak portion of the pulse in the pulling motion constituting the expansion / contraction motion (step S20).

13 and 14 are graphs showing acceleration waveforms of expansion and contraction operation. The horizontal axis shows time and the vertical axis shows acceleration. 13 and 14 are obtained, for example, when user A performs shadow boxing. 13 and 14 show changes in acceleration during operation until the fist is extended and pulled back. The acceleration on the vertical axis generally indicates the acceleration in the direction in which the fist moves.

As shown in FIG. 14, in the expansion / contraction operation, the extension operation and the pulling operation are continuously performed. The peak of the pulse that first appears at the start of the movement indicates that the arm is extended and the fist is accelerating. The acceleration of the fist is increasing in the front part of the mountain part. At the rear part of the mountain part, the acceleration decreases because the fist movable area limit is approached. After that, when the fist approaches the limit of the range of motion, the acceleration becomes a negative value and the speed of the fist slows down.
After the stretching motion is completed, the pulling motion, that is, the motion of returning the fist in the opposite direction, causes a negative acceleration (time 0.0257 to 0.0265). When the motion is changed from the stretching motion to the pulling motion, the acceleration further decreases, but it does not appear as a valley because the negative acceleration in the fist movable area is large. At the end of the pulling motion, overshoot (the peak of acceleration) occurs due to the motion of returning the fist to its original position for the next motion.

FIG. 15 is an explanatory diagram illustrating a method of calculating muscle strength strength, instantaneousness and muscle strength, and agility strength, agility and agility based on acceleration data during expansion and contraction operation. Similar to FIGS. 13 and 14, the horizontal axis represents time and the vertical axis represents acceleration. In FIG. 15, point A is the starting point of the stretching operation. Point B indicates the time when the acceleration changes from positive to negative (zero cross point). Point C is a displacement point for the stretching operation and the pulling operation. The displacement point C is a point where the acceleration changes abruptly after passing through the minimum point of the acceleration in the valley portion of the pulse in the stretching operation. Point D indicates the time when the acceleration changes from negative to positive (zero cross point). Point E is the end point of the pulling operation.

The muscle strength strength, instantaneousness, and muscle strength based on the acceleration data of the peak portion of the pulse in the stretching motion are represented by the above equations (4), (5), and (6) as in the stretching motion.
The agility intensity, agility, and agility based on the acceleration data of the valley portion of the pulse in the stretching motion are represented by the above equations (7), (8), and (9) as in the stretching motion. However, point C is a change from the stretching operation to the pulling operation, unlike the one during the stretching stop operation.

Further, the muscle strength, the instantaneousness, and the muscle strength based on the acceleration data of the valley portion of the pulse in the pulling motion are represented by the following equations (10), (11), and (12) as in the stretching stop motion. However, the signs are reversed.
Muscle strength = T10 x Σ (acceleration of pulse valley) ... (10)
Instantaneousness = T11 × (TD-TC) ... (11)
Muscle strength = T12 x Σ (acceleration of pulse valley) / (TD-TC) ... (12)
However, T10, T11 and T12 are constants for adjusting the values of muscle strength, instantaneousness and muscle strength so as to be numerical values of 2 to 4 digits.

The agility intensity, agility, and agility based on the acceleration data of the peak portion of the pulse in the stretching motion are represented by the following equations (13), (14), and (15) as in the stretching motion. However, the signs are reversed.
Agility strength = T13 x Σ (acceleration of pulse peak) ... (13)
Agility = T14 × (TE-TD) ... (14)
Agility = T15 x Σ (acceleration of pulse peak) / (TE-TD) ... (15)
However, T13, T14 and T15 are constants for adjusting the agility intensity, agility and agility values to be 2 to 4 digit values.

The control unit 20 that executes the processes of steps S14 to S20 functions as a calculation unit that calculates muscle function parameters indicating muscle function based on the acquired sensor data.

The control unit 20 that has completed the calculation of the muscle function parameters stores the calculated muscle function parameters and the date and time information measured by the clock unit in the storage unit 21 (step S21). Further, the control unit 20 may also store the sensor data together.

The control unit 20 calculates statistical values of the muscle function parameters based on the plurality of muscle function parameters (step S22). For example, the control unit 20 calculates statistical values of a plurality of muscle function parameters calculated based on a plurality of pulses included in the waveform of acceleration related to a series of movements of the exercising user A. For example, the storage unit 21 stores a plurality of muscle function parameters related to a series of movements. The statistical values of the muscle function parameters may be calculated for each type of exercise of the user. The statistical value is, for example, a value extracted or calculated according to an average value, a maximum value, a mode value, or other specific conditions of a plurality of muscle function parameters.

FIG. 16 is a schematic diagram showing a display example of muscle function parameters. As shown in FIG. 16, the control unit 20 causes the display unit 25 to display the muscle function parameters calculated this time (step S23). Further, the control unit 20 causes the display unit 25 to display one or more muscle function parameters calculated in the past together with the muscle function parameters this time (step S24). That is, the control unit 20 displays the temporal change of the muscle function parameter. For example, the control unit 20 may display the time change of impact force, muscle strength, and agility in a line graph as shown in FIG. The time change may be displayed on a daily basis, a monthly basis, or a yearly basis.
In the example of FIG. 16, impact force, muscle strength, and agility are illustrated, but muscle function parameters may be displayed for each type of movement. For example, the control unit 20 displays the impact strength, impact property, and impact force in the impact operation on the display unit 25. Further, the control unit 20 displays the muscle strength strength, the instantaneousness and the muscle strength, and the agility strength, the agility and the agility in the stretching stop operation on the display unit 25. Further, the control unit 20 displays the muscular strength, instantaneousness and muscular strength in the expansion / contraction motion, and the agility strength, agility and agility on the display unit 25.
Each muscle function parameter displayed on the display unit 25 is, for example, an average value of a series of movements, a maximum value, or a statistical value extracted from a specific condition.
The time change may be a contrast between the current value of each muscle function parameter and the value one month ago or one year ago. For example, the control unit 20 may display a bar graph showing the current value of each muscle function parameter and the value one month ago or one year ago side by side. The method of comparing and displaying the current value of each muscle function parameter and the past value is not particularly limited.

Further, as shown in FIG. 16, the control unit 20 causes the display unit 25 to display the comparison result between the calculated muscle function parameter and the predetermined muscle function parameter (step S25). The predetermined muscle function parameter is a muscle function parameter obtained when a target person, for example, an expert (instructor) of the user A performs an impact operation, an extension stop operation, an extension / contraction operation, or the like.
The mode of the comparison result is not particularly limited, and may be the difference between the muscle function parameter and the predetermined muscle function parameter, and the score of the user A based on the predetermined muscle function parameter (out of 100 points). It may be a score, etc.) or an evaluation degree (evaluation rank, etc. in a 4-grade evaluation).

FIG. 17 is a schematic diagram showing a display example of the acceleration waveform. As shown in FIG. 17, the control unit 20 causes the display unit 25 to display a comparison result between the waveform indicated by the sensor data received in the process of step S11 and the waveform indicated by the predetermined data (step S26). The predetermined data is sensor data obtained when a target person, for example, an expert (instructor) of user A performs an impact operation, an extension stop operation, and an extension / contraction operation.
When comparing the waveform indicated by the received sensor data with the waveform related to the predetermined data, a portion having a high similarity to the predetermined data is extracted from the waveform indicated by the sensor data, and the extracted sensor data indicates the portion. It is advisable to compare the waveform with the waveform indicated by the predetermined data. That is, it is preferable to compare the waveforms when the user A and the expert (instructor) perform the same operation.
Further, when comparing the waveforms, the muscle function parameter calculation device 2 does not compare the magnitude of the peak value and the pulse width, compares the timing at which the pulsed waveform appears, and displays or outputs the comparison result. You can do it.

<Type of exercise suitable for muscle function parameter calculation system>
One of the practical purposes of the muscle function parameter calculation system according to the present embodiment is to arouse the consciousness for health promotion, that is, the consciousness for exercising. For example, an active senior who emphasizes health can monitor the improvement of his / her muscle function by exercise by using the muscle function parameter calculation system.
Exercises performed in a sports gym are appropriate exercises for assessing muscle function. Appropriate exercise is as follows.

Using treadmills, bikes, elliptical trainers, and step machines, cardio-type exercise aimed at raising the heart rate and burning body fat is because the moving parts (arms, legs, etc.) move actively. , Suitable for measuring muscle function.

Dance-type exercises such as Zumba and aerobics are suitable for measuring muscle function because they are training to move the entire body in time with music.

Martial arts exercises such as kickboxing, taekwondo, karate, Muay Thai, and boxing are suitable for measuring muscle function because they move the moving parts (arms, legs, etc.) instantaneously.

Running driving such as 100m dash is suitable for measuring muscle function because it is an exercise having instantaneousness that maximizes muscle strength.

Running or walking can also be a measure of muscle function. For running or walking, muscle ability can be evaluated by a relative value evaluation with an expert (instructor) or oneself in the past.
In the case of running or walking, the user A runs strongly, runs weakly, or takes a break. Therefore, the detected acceleration includes acceleration data unsuitable for measuring muscle function. In this case, as described in the third embodiment described later, it is preferable to extract only the acceleration data suitable for the measurement or evaluation of the muscle function by using the AI (learning model).

Various athletics, swimming, gymnastics, bicycle sports, martial arts / fighting, shooting sports such as darts, ball sports, skiing, snowboarding, skating, athletics, water sports such as surfing, snow sports such as snow battles, land such as roller skating The present embodiment can be applied to gliding sports, competitive sports, and any other exercise with instantaneous movements.

Further, the muscle strength parameter, the agility parameter, and the like may be calculated for a plurality of types of exercises, and the muscle function of the user A may be comprehensively evaluated. For example, if user A is conducting a gym kickboxing lesson, a dance lesson, or a running, the plurality of them are comprehensively judged and evaluated.

<Combined use with other sensors>
It is possible to obtain more value-added information by using the body sensing data obtained from other sensors such as respiration, pulse, and blood pressure and the muscle function parameters obtained by the muscle function parameter calculation system according to the present embodiment. can. For example, even if the exercise has the same muscular strength performance, if the heart rate and the respiratory rate do not increase, it is possible to evaluate that the heart rate and the respiratory function ability are improved. The comparison target may be a third party other than user A, or may be past user A.

<Number of wearable devices installed and parts to be installed>
Further, in the first embodiment, an example in which one wearable device 1 is used has been described, but a plurality of wearable devices 1 may be used. In this case, it is preferable to synchronize the times of the plurality of wearable devices 1.
By using the plurality of wearable devices 1, it is possible to measure or evaluate the muscle function of the entire body of the user A. The user A may wear the wearable device 1 on both hands, both arms, and both feet, for example. By comparing the movements of the left and right hands, arms, and legs, the balance of left and right muscle functions can be grasped.
Further, the user A may wear the wearable device 1 on the waist and the back. It is possible to grasp the muscle function of the trunk or the whole body.
When performing relative value comparison, the user A or the expert (instructor) to be compared needs to wear the wearable device 1 on the same part.

According to the computer program 21a, the muscle function parameter calculation device 2, the muscle function parameter calculation system, and the muscle function parameter calculation method according to the first embodiment configured in this way, it is based on the sensor data indicating the acceleration acquired from the wearable device 1. Therefore, the muscle function parameter indicating the muscle function of the user A can be calculated.

Many systems have been developed to monitor the condition of the human body. For example, by monitoring respiration, pulse, blood pressure, etc., the amount of activity and calories burned are measured to raise awareness of exercise promotion such as how much exercise has been performed. On the other hand, we measure the posture in sports by measuring the angle, angular velocity, speed, and acceleration of each part of the body, and we are trying to get closer to the ideal posture.
The former is only a means of quantifying how much exercise you have done for daily exercise, not an index of how much your muscle strength is. The latter also improves the ability for various sports by quantifying and improving the posture state, but it does not monitor the potential ability in the first place.
The muscle function parameter calculation system according to the present embodiment can monitor potential muscle functions that are difficult to monitor by the above method. User A can grasp the level of his / her muscle function as compared with others. User A can grasp how much the muscle function has improved before and after the continuous exercise.

According to the muscle function parameter calculation device 2 and the like, the calculated muscle function parameters such as muscle strength and agility can be displayed or output. User A can check the displayed muscle function parameters and perform health management.

According to the muscle function parameter calculation device 2 and the like, it is possible to calculate muscle function parameters such as impact strength, impact property, and impact force in the impact motion of the user A.

According to the muscle function parameter calculation device 2 and the like, it is possible to calculate muscle function parameters such as muscle strength, instantaneousness, and muscle strength related to the muscle strength of the user A.

According to the muscle function parameter calculation device 2 and the like, it is possible to calculate muscle function parameters such as agility strength, agility and agility related to the agility of the user A.

According to the muscle function parameter calculation device 2 or the like, the statistical value of the muscle function parameter calculated based on a plurality of pulses included in the waveform of the acceleration related to the series of movements of the exercising user A is calculated, and the statistical value of the muscle function parameter is calculated. Can be displayed on the display unit 25.

According to the muscle function parameter calculation device 2 and the like, the types of movements of the user A can be classified into at least impact movements, extension / contraction movements, and expansion / contraction movements, and muscle function parameters can be calculated according to the types of the classified movements. can. The pulse shape of acceleration differs depending on the type of operation. By classifying the types of movements and calculating the muscle function parameters, the muscle strength parameters can be calculated more accurately.

According to the muscle function parameter calculation device 2 and the like, the calculated impact strength, impact and impact force, muscle strength, instantaneousness and muscle strength, and muscle function parameters such as agility strength, agility and agility are stored in the storage unit 21. Can be memorized in.

According to the muscle function parameter calculation device 2 and the like, it is possible to display or output the temporal change of the muscle function parameter of the user A. By displaying the changes in the muscle function parameters, the motivation for health management can be given to the user A. In addition, the user A can grasp the increase in muscle strength, the decrease in muscle strength, the increase in physical strength, the decrease in physical strength, and the like.

According to the muscle function parameter calculation device 2 and the like, it is possible to display or output the comparison result between the muscle function parameter of the user A and the muscle function parameter of the expert (instructor). By displaying the comparison result with the muscle function parameter of the expert (instructor), the user A can confirm his / her progress and can give the user A motivation for health management.

According to the muscle function parameter calculation device 2 and the like, it is possible to display or output the comparison result between the waveform of the sensor data related to the user A and the waveform of the sensor data related to the expert (instructor). By displaying the comparison result with the sensor data waveform of an expert (instructor), the degree of improvement of oneself can be confirmed, and the motivation for health management can be given to the user A.

By setting the detection range of the acceleration sensor 11 to 20 [G], the wearable device 1 can detect the movement of the user A who performs a movement without impact while suppressing a decrease in measurement accuracy.

By setting the detection range of the acceleration sensor 11 to 40 [G], the wearable device 1 can detect the movement of the user A who performs a movement accompanied by an impact while suppressing a decrease in measurement accuracy.

Since the wearable device 1 is configured to calculate the scalar value of the acceleration and wirelessly transmit it to the muscle function parameter calculation device 2, it is possible to reduce the amount of data to be wirelessly transmitted and the calculation processing load of the muscle function parameter calculation device 2. .. The battery consumption of the wearable device 1 can be suppressed.

The wearable device 1 has a configuration in which sensor data related to acceleration below the threshold value is removed and sensor data related to acceleration above the threshold value is wirelessly transmitted to the muscle function parameter calculation device 2, and unnecessary data transmission can be avoided. .. The battery consumption of the wearable device 1 can be suppressed.

The wearable device 1 can transmit sensor data indicating the acceleration generated by the movement of the user A to the muscle function parameter calculation device 2 by removing the gravity component from the acceleration detected by the acceleration sensor 11. Therefore, the muscle function parameters can be calculated more accurately.

In the first embodiment, the wearable device 1 has described an example in which the sensor data is constantly and periodically wirelessly transmitted to the muscle function parameter calculation device 2, but the detected sensor data is accumulated in a certain amount and the accumulated sensor. The data may be configured to be collectively transmitted wirelessly.
Further, the wearable device 1 may be configured so that the memory card can be attached and detached and the sensor data is stored in the memory card. The muscle function parameter calculation device 2 can also be provided with a removable memory card, and the muscle function parameter calculation device 2 reads sensor data from the mounted memory card, and is described in the first embodiment based on the read sensor data. The muscle function parameters may be calculated as described above.
Further, the wearable device 1 may transmit the sensor data accumulated by NFC (Near field communication) to the muscle function parameter calculation device 2.
Further, in the configuration in which the wearable device 1 accumulates the sensor data and the muscle function parameter calculation device 2 collectively acquires the sensor data from the wearable device 1, the wearable device 1 only detects an acceleration exceeding a predetermined threshold value. , The sensor data may be configured to be wirelessly transmitted to the muscle function parameter calculation device 2. The power consumption of the wearable device 1 and the muscle function parameter calculation device 2 can be suppressed.

Further, in the first embodiment, the example in which the muscle function parameter calculation device 2 which is a wireless communication device calculates the muscle function parameter of the user A has been described, but the server 3 may calculate the muscle function parameter. The wireless communication device having the same hardware configuration as the muscle function parameter calculation device 2 acquires acceleration data from the wearable device 1 and wirelessly transmits the acquired acceleration data to the server 3. The server 3 calculates the muscle function parameters by executing the computer program 21a according to the first embodiment, and transmits the calculated muscle function parameters to the wireless communication device. The wireless communication device acquires the muscle function parameters calculated by the server 3 and displays them on the display unit 25. In addition, various functions described in the first embodiment may be appropriately executed by the server 3 and the execution result may be displayed on the wireless communication terminal.

(Embodiment 2)
The computer program 21a, the muscle function parameter calculation device 2, the muscle function parameter calculation system, and the muscle function parameter calculation method according to the second embodiment differ mainly from the first embodiment in the calculation method of the muscle function parameters. The points will be explained. Since other configurations and actions and effects are the same as those in the first embodiment, the corresponding parts are designated by the same reference numerals and detailed description thereof will be omitted.

FIG. 18 is a functional block diagram of the muscle function parameter calculation device 2 according to the second embodiment. The muscle function parameter calculation device 2 according to the second embodiment includes a sensor data acquisition unit 20a, a pulse selection unit 20b, a first learning model 20c, a second learning model 20d, and a third learning model 20e. When executing machine learning processing, the control unit 20 of the muscle function parameter calculation device 2 according to the second embodiment has a GPU (Graphics Processing Unit), a GPGPU (General-purpose computing on graphics processing units), and a TPU (Tensor Processing Unit). The configuration having the above is preferable.

The sensor data acquisition unit 20a acquires sensor data from the wearable device 1 and gives the acquired sensor data to the pulse selection unit 20b.

The pulse selection unit 20b selects the sensor data of the impact operation from the acquired sensor data, and outputs the sensor data of the pulse related to the impact operation to the first learning model 20c.
Further, the pulse selection unit 20b selects the sensor data related to the de-stretching operation from the acquired sensor data, and outputs the sensor data related to the de-stretching operation to the second learning model 20d.
Further, the pulse selection unit 20b outputs the sensor data related to the expansion / contraction operation among the acquired sensor data to the third learning model 20e.
The selection of the sensor data by the pulse selection unit 20b can be performed by, for example, pattern matching of the acceleration waveform represented by the sensor data. As shown in FIG. 6, the waveform of the acceleration related to the impact operation is composed of a mountain-shaped pulse shape and a valley-shaped pulse shape following the mountain-shaped pulse shape. It is possible to discriminate the sensor data of. Further, as shown in FIG. 9, the waveform of the acceleration related to the stretching stop operation is composed of a mountain-shaped pulse shape, a valley-shaped pulse shape following the mountain-shaped pulse shape, and a small mountain-shaped pulse shape following the peak-shaped pulse shape. The selection unit 20b can determine the sensor data of the stretch stop operation by extracting such a feature. Further, as shown in FIG. 13, the waveform of the acceleration related to the expansion / contraction operation is a mountain-shaped pulse shape, a valley-shaped pulse shape following the mountain-shaped pulse shape, and a mountain-shaped pulse shape following the mountain-shaped pulse shape (the first mountain-shaped pulse). Since the pulse selection unit 20b has the same size as the shape), the pulse selection unit 20b can determine the sensor data of the expansion / contraction operation by extracting such a feature.

The first learning model 20c is a neural network having an input layer for inputting sensor data related to impact motion, an intermediate layer, and an output layer for outputting muscle function parameters such as impact strength, impact property, and impact force. .. The first learning model 20c is provided with, for example, a recurrent neural network (RNN), and when time-series sensor data is input, the impact strength, impact property, and impact force of user A are output. It is good to train RNN. Further, the first learning model 20c may be configured to process the waveform indicated by the sensor data as an image. The first learning model 20c has a convolution layer and a pooling layer (not shown) for processing the image data of the waveform indicated by the sensor data.
The method of creating the first learning model 20c is as follows. First, learning data having sensor data related to a plurality of impact movements and muscle function parameters such as impact strength, impact property, and impact force, which are evaluations for each sensor data, is prepared. Then, sensor data is input to the neural network before learning, and various methods such as gradient descent and backpropagation are used so that the error between the output muscle function parameters and the muscle function parameters of the training data becomes small. Optimize the parameters and let the neural network machine learn. By repeating such machine learning, the first learning model 20c can be generated.

The second learning model 20d sets the input layer into which the sensor data related to the stretching stop motion is input, the intermediate layer, the muscle strength strength, the instantaneousness and the muscle strength, and the muscle function parameters such as the agility strength, the agility and the agility. It is a neural network having an output layer for output. The configuration of the second learning model 20d is the same as that of the first learning model 20c. The method of creating the second learning model 20d is as follows. First, learning data having a plurality of sensor data related to the extension stop motion, muscle strength strength, instantaneousness and muscle strength, which are evaluations for each sensor data, and muscle function parameters such as agility strength, agility and agility. prepare. Then, the sensor data is input to the neural network before learning, and the gradient descent method, error backpropagation method, etc. are used so that the error between the output muscle function parameter and the muscle function parameter of the training data becomes small. Optimize various parameters and let the neural network machine learn. By repeating such machine learning, the second learning model 20d can be generated.

The third learning model 20e outputs an input layer in which sensor data related to expansion and contraction is input, an intermediate layer, muscle strength strength, instantaneousness and muscle strength, and muscle function parameters such as agility strength, agility and agility. It is a neural network having an output layer and an output layer. The configuration of the third learning model 20e is the same as that of the first learning model 20c. The method of creating the third learning model 20e is as follows. First, learning data having pulse sensor data related to a plurality of expansion and contraction movements, muscle strength strength, instantaneousness and muscle strength, which are evaluations for each sensor data, and muscle function parameters such as agility strength, agility and agility are obtained. prepare. Then, the sensor data is input to the neural network before learning, and the gradient descent method, error backpropagation method, etc. are used so that the error between the output muscle function parameter and the muscle function parameter of the training data becomes small. Optimize various parameters and let the neural network machine learn. By repeating such machine learning, the third learning model 20e can be generated.

The same effect as that of the first embodiment is obtained in the computer program 21a, the muscle function parameter calculation device 2, the muscle function parameter calculation system, and the muscle function parameter calculation method according to the second embodiment.

The structure of the first learning model 20c, the second learning model 20d, the number of intermediate layers of the third learning model 20e, the number of neurons in each layer, and the like are not particularly limited, and the types of neural networks are also particularly limited. It is not something that is done.
Further, in the second embodiment, a neural network is shown as an example of a machine learning device that calculates various muscle function parameters, but a known trained model such as an SVM (Support Vector Machine), a Bayesian network, or a regression tree is used. It may be used to learn the relationship between the sensor data and the muscle function parameters.

Further, in the second embodiment, the sensor data is classified into the sensor data related to the impact motion, the sensor data related to the stretching motion, and the sensor data of the stretching motion, and the first learning model 20c, the first learning model 20c, which analyzes each sensor data. The 2 learning model 20d and the 3rd learning model 20e have been described, but the sensor data related to the pulling motion is further classified, and based on the sensor data, the muscle strength, the instantaneousness, the muscle strength, the agility strength, the agility and the agility A learning model for calculating force may be provided.

In the second embodiment, an example in which the muscle function parameter calculation device 2 which is a wireless communication device calculates various muscle function parameters using the learning model has been described, but the server 3 is configured to calculate the muscle function parameters. You may.
The learning of the neural network may also be executed by the wireless communication device or the server 3.

(Embodiment 3)
The computer program 21a, the muscle function parameter calculation device 2, the muscle function parameter calculation system, and the muscle function parameter calculation method according to the third embodiment use a learning model to determine the acceleration waveform obtained from the acceleration sensor 11 and the type of exercise. Since the point of recognizing the relationship is different from that of the first or second embodiment, the above difference will be mainly described below. Since other configurations and actions and effects are the same as those of the first or second embodiment, the corresponding parts are designated by the same reference numerals and detailed description thereof will be omitted.

FIG. 19 is a functional block diagram of the muscle function parameter calculation device 2 according to the third embodiment. The muscle function parameter calculation device 2 according to the third embodiment includes a fourth learning model 20f. When executing machine learning processing, the control unit 20 of the muscle function parameter calculation device 2 according to the third embodiment has a GPU (Graphics Processing Unit), a GPGPU (General-purpose computing on graphics processing units), and a TPU (Tensor Processing Unit). The configuration having the above is preferable.

In the fourth learning model 20f, information indicating a plurality of accelerations and measurement time points included in the sensor data, for example, an input layer and an intermediate layer in which acceleration waveforms are input, and the input accelerations are applied to a specific motion of the user A. It is a neural network having an output layer for outputting information for determining whether or not the information indicates a pulse of acceleration based on the acceleration. The fourth learning model 20f is provided with, for example, a recurrent neural network, and when time-series sensor data is input, it is preferable to train the RNN so as to output information indicating the type of movement performed by the user A. .. Types of exercise include impact exercise, stretch-stopping exercise, and stretching exercise. Further, the fourth learning model 20f may be configured to process the waveform indicated by the sensor data as an image. The fourth learning model 20f has a convolution layer and a pooling layer (not shown) for processing the image data of the waveform indicated by the sensor data. Further, the fourth learning model 20f may be configured to discriminate the pulling motion.
The method of creating the fourth learning model 20f is as follows. First, learning data having a plurality of sensor data generated when the user A performs a specific exercise and the type of the exercise (teacher data) is prepared. The learning data may include learning data in which sensor data that does not correspond to any exercise and teacher data indicating that the sensor data is not during exercise are associated with each other. Then, the sensor data is input to the neural network before learning, and the neural network is machine-learned so that the error between the output type of motion and the type of motion indicated by the training data becomes small. By repeating such machine learning, the fourth learning model 20f can be generated.

The control unit 20 of the muscle function parameter calculation device 2 causes the fourth learning model 20f to input sensor data, and identifies the type of exercise performed by the user A based on the information output from the fourth learning model 20f. can do. For example, the control unit 20 identifies that the user A is performing an impact operation, an extension stop operation, or an extension / contraction operation. Then, the control unit 20 extracts information indicating the acceleration corresponding to the specified motion and the measurement time point, and removes the sensor data not corresponding to the motion. The control unit 20 calculates muscle function parameters based on the acceleration of the movement and the information at the time of measurement.

According to the third embodiment configured in this way, the control unit 20 can recognize the type of the exercise performed by the user A and extract the sensor data corresponding to the exercise, so that the muscle of the user A can be extracted. The muscle function parameters representing the function can be calculated more accurately and accurately.

In the third embodiment, a neural network is shown as an example of a machine learning device that recognizes the type of motion, but a known learning model such as an SVM (Support Vector Machine), a Bayesian network, or a regression tree is used. Then, the relationship between the acceleration waveform and the type of motion may be learned.

In the third embodiment, an example in which the muscle function parameter calculation device 2 which is a wireless communication device specifies the type of exercise by using the learning model has been described, but the server 3 side specifies the type of exercise of the user A. It may be configured as.
The learning of the neural network may also be executed by the wireless communication device or the server 3.

Further, in the first to third embodiments, an example of calculating the muscle function parameter based on the sensor data at the time of a specific operation such as sandback striking has been described, but the muscle function parameter calculation device 2 is a basic operation such as walking. It may be configured to evaluate the muscle function of the user A based on the sensor data of the time. The muscle function parameter calculation device 2 detects the walking motion of the user A, and when the walking motion is detected, evaluates the muscle function of the user A based on the sensor data for a predetermined time, and displays or outputs the evaluation value. good. User A can confirm the displayed evaluation value and perform health management.

Further, although an example of basically calculating the muscle function parameter based on the sensor data of one movement has been described, the muscle function parameter and other evaluation values are calculated based on the sensor data acquired in a certain period of time. It may be configured to do so.
For example, the muscle function parameter calculation device 2 may evaluate the efforts of the user A based on the acceleration data for a certain period of time, and display or output the evaluation value.
Further, the muscle function parameter calculation device 2 may estimate the potential muscle strength, impact force, and agility of the user A based on the acceleration data for a certain period of time, and display or output the estimation result.
Further, the muscle function parameter calculation device 2 may calculate the maximum muscle force, impact force, and agility of the user A based on the acceleration data for a certain period of time, and display or output the calculation result.

1 Wearable device 2 Muscle function parameter calculation device 3 Server 11 Accelerometer 12 Transmission unit 20 Control unit 20a Sensor data acquisition unit 20b Pulse selection unit 20c 1st learning model 20d 2nd learning model 20e 3rd learning model 20f 4th learning model 21 Storage unit 21a Computer program 22 Wireless communication unit 23 Short-range communication unit 24 Vibrator 25 Display unit 26 Operation unit 27 Speaker 28 Microphone A User N Communication network

Claims (20)

Obtain sensor data from a wearable device worn on the user's body and
A computer program for causing a computer to execute a process of calculating at least muscle function parameters indicating the level of muscle function of a user based on the acquired sensor data. The computer program according to claim 1, wherein a computer executes a process of displaying the calculated muscle function parameters on a display unit. The sensor data is acquired from the wearable device having an acceleration sensor, and the sensor data is acquired.
A claim for causing a computer to perform a process of calculating the muscle function parameters related to the impact force generated by the user's impact operation based on the information of the valley portion of the pulse included in the waveform indicated by the sensor data acquired during the impact operation. 1 or the computer program according to claim 2. The sensor data is acquired from the wearable device having an acceleration sensor, and the sensor data is acquired.
Claims 1 to 1 for causing a computer to execute a process of calculating the muscle function parameter related to the user's muscle strength based on the information of the peak portion of the pulse included in the waveform indicated by the sensor data acquired during the stretch stop operation. The computer program according to any one of 3. The sensor data is acquired from the wearable device having an acceleration sensor, and the sensor data is acquired.
A claim for causing a computer to perform a process of calculating the muscle function parameter related to the user's agility based on the information of the valley portion following the peak portion of the pulse included in the waveform indicated by the sensor data acquired during the stretch stop operation. The computer program according to any one of claims 1 to 4. The sensor data is acquired from the wearable device having an acceleration sensor, and the sensor data is acquired.
From claim 1 for causing a computer to execute a process of calculating the muscle function parameter related to the user's muscle strength based on the information of the peak portion of the pulse in the stretching operation included in the waveform indicated by the sensor data acquired during the stretching operation. The computer program according to any one of claims 5. The sensor data is acquired from the wearable device having an acceleration sensor, and the sensor data is acquired.
To make the computer execute the process of calculating the muscle function parameter related to the agility of the user based on the information of the valley portion following the peak portion of the pulse in the stretching motion included in the waveform indicated by the sensor data acquired during the stretching motion. The computer program according to any one of claims 1 to 6. The sensor data is acquired from the wearable device having an acceleration sensor, and the sensor data is acquired.
From claim 1 for causing a computer to execute a process of calculating the muscle function parameter related to the user's muscle strength based on the information of the valley portion of the pulse in the pulling operation included in the waveform indicated by the sensor data acquired during the expansion / contraction operation. The computer program according to any one of claims 7. The sensor data is acquired from the wearable device having an acceleration sensor, and the sensor data is acquired.
To make the computer execute the process of calculating the muscle function parameter related to the agility of the user based on the information of the peak portion following the valley portion of the pulse in the pulling motion included in the waveform indicated by the sensor data acquired during the expansion / contraction motion. The computer program according to any one of claims 1 to 8. A plurality of the muscle function parameters are calculated based on the plurality of pulses included in the waveform indicated by the sensor data.
The computer program according to any one of claims 1 to 9, which calculates the calculated statistical values of a plurality of muscle function parameters. The sensor data is acquired from the wearable device having an acceleration sensor, and the sensor data is acquired.
Based on the acquired sensor data, the types of motion are classified into at least impact motion, stretch stop motion, and stretch motion.
The computer program according to any one of claims 1 to 10, for causing a computer to execute a process of calculating the muscle function parameters according to the type of classified motion. The computer program according to any one of claims 1 to 11, for causing a computer to execute a process of storing the calculated muscle function parameters in a storage unit. The computer program according to claim 12, wherein the computer is made to execute a process of outputting a temporal change of the muscle function parameter stored in the storage unit. The computer program according to any one of claims 1 to 13, for causing a computer to execute a process of outputting a comparison result between the calculated muscle function parameter and a predetermined muscle function parameter. The computer program according to any one of claims 1 to 14, for causing a computer to execute a process of outputting a comparison result between the waveform indicated by the sensor data and the waveform indicated by the predetermined data. The computer program according to any one of claims 1 to 15, for causing a computer to execute a process of removing a non-pulse waveform included in the sensor data as noise. It is provided with a learning model trained so that when the sensor data is input, the muscle function parameters are output.
The computer program according to any one of claims 1 to 16, for causing a computer to execute a process of inputting the acquired sensor data into the learning model and outputting the muscle function parameters from the learning model. .. An acquisition unit that acquires sensor data from a wearable device worn on the user's body,
A muscle function parameter calculation device including a calculation unit that calculates at least a muscle function parameter indicating the level of the user's muscle function based on the sensor data acquired by the acquisition unit. Wearable devices with accelerometers and
Calculation to calculate at least muscle function parameters indicating the level of muscle function of the user based on the acquisition unit that acquires sensor data from the wearable device attached to the user's body and the sensor data acquired by the acquisition unit. A muscle function parameter calculation system including a muscle function parameter calculation device having a part. Obtain sensor data from a wearable device worn on the user's body and
A muscle function parameter calculation method for calculating at least a muscle function parameter indicating the level of a user's muscle function based on the acquired sensor data.

Images