Evaluation index calculation device, method, and program

The device calculates an evaluation index by analyzing electromyogram amplitude features and normalizing muscle activity to assess motor skills effectively for isometric exercises, addressing the challenge of constant muscle activity intensity and improving skill evaluation.

WO2024236780A1PCT designated stage expired Publication Date: 2024-11-21NT T INC
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
PCT/JP2023/018469
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-05-17
Publication Date
2024-11-21

AI Technical Summary

Technical Problem

Existing methods for evaluating exercise skills, such as isometric movements where muscles hold an object, fail to accurately measure steady muscle activity over time due to constant muscle activity intensity and difficulty in converging analysis results, as they are based on changing activity intensity and time-series analysis.

Method used

A device and method that calculates an evaluation index by measuring the amplitude features of electromyograms, normalizing muscle activity, determining relative muscle activity strength, and calculating the bias in muscle activity distribution over time to assess motor skills effectively.

Benefits of technology

Enables appropriate evaluation of motor skills for holding objects by adjusting muscle activity intensity, reducing fatigue, and providing a suitable skill evaluation index for isometric exercises.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure JP2023018469_21112024_PF_FP_ABST
    Figure JP2023018469_21112024_PF_FP_ABST
Patent Text Reader

Abstract

This evaluation index calculation device according to one embodiment comprises: a first calculation unit that calculates an amplitude characteristic amount of a time-series electromyography of the muscles of a subject; a second calculation unit that calculates the relative muscle activity amount of each muscle at different times, on the basis of a value obtained by normalizing the amplitude characteristic amount calculated by the first calculation unit; a third a calculation unit that calculates information indicating the strength / weakness of the calculated relative muscle activity amount, on the basis of the relative muscle activity amount calculated by the second calculation unit and a relative muscle activity amount in a reference interval; a fourth calculation unit that calculates a distribution of the time when there is an instance of strength / weakness of the relative muscle activity, the strength / weakness being indicated by the information calculated by the third calculation unit; and a fifth calculation unit that calculates the bias of the activity amount of the muscles of the subject, on the basis of the distribution of the time when there is an instance of strength / weakness, the distribution of time having been calculated by the fourth calculation unit.
Need to check novelty before this filing date? Find Prior Art

Description

Evaluation index calculation device, method, and program

[0001] An embodiment of the present invention relates to an evaluation index calculation device, method, and program.

[0002] For example, a person may hold a box in both hands for a certain period of time, but this movement involves almost no movement of the joints or body parts, making it a static movement. Such static movements, involving almost no movement of the joints or body parts, have long been analyzed as isometric movements, and the muscle activity in such movements does not change significantly over time. For this reason, the ratio of muscle activities involved in performing the movement is used to evaluate the skill of the movement, as disclosed in, for example, Non-Patent Document 1.

[0003] Furthermore, exercises such as walking can be considered dynamic because they involve movements of joints and body parts. During such exercises, muscle activity varies significantly over time, resulting in significant changes in the intensity of muscle activity. For this reason, dimension reduction techniques are used to evaluate exercise skill, such as the non-negative matrix factorization (NMF) algorithm disclosed in Non-Patent Document 2, which expresses the time-series activity of multiple muscles involved in performing a movement using a small number of combinations.

[0004] Park, JH, Chung, SW, Lee, SJ, Lee, JW, & Oh, KS (2020). Evaluation of the electromyographic amplitude-to-work ratio in the infraspinatus muscle during external shoulder rotation exercises: a comparison of concentric isotonic and isokinetic exercises. Orthopedic Journal of Sports Medicine, 8(7), 2325967120932459.Tresch, Matthew C., Vincent CK Cheung, and Andrea d'Avella. "Matrix factorization algorithms for the identification of muscle synergies: evaluation on simulated and experimental data sets." Journal of neurophysiology 95.4 (2006): 2199-2212.

[0005] When humans continue to hold an object, there is an isometric movement aspect in which joints and body parts do not move significantly. However, when using the muscles of both arms, for example, there are multiple combinations of muscles that humans use to hold an object, and it is thought that there is an inherent change in muscle activity over time.

[0006] When assessing the skill of such exercises, the approach using the ratio of multiple muscle activities involved in performing the exercise, as disclosed in Non-Patent Document 1, assumes that muscle activity is steady and does not change over time. However, in the exercise of a person continuously holding an object, although there is steady muscle activity, the activity intensity of each muscle changes over time, so the above approach cannot be said to be suitable for skill assessment.

[0007] Furthermore, the analysis using the dimension reduction method disclosed in Non-Patent Document 2 is based on the assumption that the strength of activity of each muscle changes significantly over time. However, in the movement of a human continuously holding an object, which is the subject of this study, there is a possibility that the solution will not converge because, while the strength of activity of each muscle changes over time, steady muscle activity also exists.

[0008] This invention was made in light of the above circumstances, and its purpose is to provide an evaluation index calculation device, method, and program that can appropriately evaluate a person's motor skills in continuing to hold an object.

[0009] An evaluation index calculation device according to one aspect of the present invention includes: a first calculation unit that calculates amplitude features of a time-series electromyogram of each muscle of a subject; a second calculation unit that calculates a relative muscle activity amount of each muscle at each time point based on normalized values ​​of the amplitude features calculated by the first calculation unit; a third calculation unit that calculates information indicating the strength of the calculated relative muscle activity amount based on the relative muscle activity amount calculated by the second calculation unit and the relative muscle activity amount in a reference interval; a fourth calculation unit that calculates a distribution of times during which the strength and weakness are occurring, which is indicated by the information indicating the strength and weakness of the relative muscle activity amount calculated by the third calculation unit; and a fifth calculation unit that calculates a bias in the activity amount of each muscle of the subject based on the distribution of times during which the strength and weakness are occurring calculated by the fourth calculation unit.

[0010] An evaluation index calculation method according to one aspect of the present invention is a method performed by an evaluation index calculation device, and includes: a first calculation unit of the evaluation index calculation device calculates amplitude feature values ​​of a time-series electromyogram of each muscle of a subject; a second calculation unit of the evaluation index calculation device calculates relative muscle activity amounts of the muscles at each time point based on normalized values ​​of the amplitude feature values ​​calculated by the first calculation unit; a third calculation unit of the evaluation index calculation device calculates information indicating the strength of the calculated relative muscle activity amounts based on the relative muscle activity amounts calculated by the second calculation unit and the relative muscle activity amounts in a reference interval; a fourth calculation unit of the evaluation index calculation device calculates a distribution of times during which the strength and weakness are occurring, indicated by the information indicating the strength and weakness of the relative muscle activity amounts calculated by the third calculation unit; and a fifth calculation unit of the evaluation index calculation device calculates a bias in the activity amounts of each muscle of the subject based on the distribution of times during which the strength and weakness are occurring calculated by the fourth calculation unit.

[0011] According to the present invention, it is possible to appropriately evaluate a person's motor skill of continuing to hold an object.

[0012] Fig. 1 is a diagram showing an application example of a skill evaluation index calculation device according to an embodiment of the present invention. Fig. 2 is a flowchart showing an example of a procedure of processing operations of the skill evaluation index calculation device. Fig. 3 is a block diagram showing an example of the hardware configuration of the skill evaluation index calculation device according to an embodiment of the present invention.

[0013] An embodiment of the present invention will be described below with reference to the drawings. In exercises where a person holds an object for a certain period of time, it is considered more suitable for long-term holding than for reducing muscle fatigue by adjusting the strength of activity of multiple muscles rather than constantly activating the muscles. The skill evaluation index calculation device according to this embodiment calculates an index for skill evaluation of such exercise.

[0014] The skill evaluation index calculation device first measures a surface electromyogram to measure muscle activity during an exercise in which a human subject holds an object, and then calculates amplitude features after removing noise from the surface electromyogram. For example, the skill evaluation index calculation device measures all muscle activity during an exercise performed over a certain period of time and calculates amplitude features.

[0015] The skill evaluation index calculation device performs normalization processing on the amplitude feature values ​​to compare between muscles, and calculates the relative muscle activity (sometimes referred to as relative muscle activity) for each of all observed muscles at each time point. This relative muscle activity can be calculated as the ratio of the amplitude feature value of the target muscle to the sum of the amplitude feature values ​​of all muscles.

[0016] In order to determine the change in the strength of muscle activity over time, the skill evaluation index calculation device defines a movement interval, which is a reference time, and obtains the average and standard deviation of the relative muscle activity of each muscle during this interval.

[0017] If the calculated relative muscle activity value for each muscle exceeds "mean + standard deviation," the skill evaluation index calculation device calculates the muscle activity strength label indicating the strength state of the calculated relative muscle activity as "strong," indicating a relatively large strength state. If the calculated relative muscle activity value is below "mean - standard deviation," the skill evaluation index calculation device calculates the muscle activity strength label of the calculated relative muscle activity as "weak," indicating a relatively small strength state. In all other cases, i.e., if the calculated relative muscle activity value is equal to or less than "mean + standard deviation" or equal to or more than "mean - standard deviation," the skill evaluation index calculation device calculates the muscle activity strength label of the calculated relative muscle activity as "average," between "strong" and "weak."

[0018] The skill evaluation index calculation device aggregates the muscle activity strength labels of "strong," "normal," and "weak" obtained for each muscle over time, calculates the time proportions that represent the proportions of time that each of the muscle activity strength labels of "strong," "normal," and "weak" occurred relative to the total time spent performing the exercise, and calculates the bias in these proportions as an index for skill evaluation of the exercise in which a person continues to hold an object.

[0019] Next, as an example, an example will be shown in which muscle activity is measured from a muscle site M of a human and an index for skill evaluation of a movement in which the human continues to hold an object for a certain period of time is calculated. As shown in Fig. 1, a skill evaluation index calculation device 100 according to this embodiment includes a surface electromyogram measurement unit 10, a normalization parameter storage unit 20, a normalization unit 30, a relative muscle activity calculation unit 40, a muscle activity strength change calculation unit 50, a time-series strength distribution calculation unit 60, and a strength bias calculation unit 70.

[0020] 2 is a flowchart showing an example of the procedure of the processing operation of the skill evaluation index calculation device. (Surface electromyogram measurement unit 10) The surface electromyogram measurement unit 10 measures time-series surface electromyogram signals m from electrodes attached to M muscles of a human subject. k (t) (k = 1, ..., M) is measured (S1). t represents time, and the surface electromyogram is measured from time 0 to T.

[0021] The surface electromyogram measuring unit 10 measures the electromyogram signal m k To remove noise during exercise, a band-pass filter that passes through a band of 20 to 500 [Hz] is applied to (t). After this application, the surface electromyogram measurement unit 10 generates the electromyogram signal m k The amplitude feature is calculated by applying a low-pass filter that passes through a frequency band of 10 [Hz] or less to the signal (t). k (t) (k = 1, ..., M). The surface electromyogram measurement unit 10 calculates the amplitude feature a(t) (= [a k (t)] (k = 1, ..., M) to the normalization unit 30.

[0022] (Normalization unit 30) The normalization unit 30 calculates the amplitude feature a(t) input from the surface electromyogram measurement unit 10 and the normalization parameter b=([b1, ..., b M ]).

[0023] The normalization unit 30 normalizes the amplitude feature a(t) at each time by the normalization parameter b to obtain a feature n k(t) (= a k (t) / b k (k = 1, ..., M)) is calculated (S2), and this feature quantity n k (t) is output to the relative muscle activity calculation unit 40.

[0024] (Normalization Parameter Storage Unit 20) The normalization parameter storage unit 20 stores normalization parameters for normalizing the amplitude feature amount of the surface electromyogram measured by the surface electromyogram measurement unit 10.

[0025] As an example of a parameter for normalizing the amplitude feature, the amplitude feature of the surface electromyogram of each muscle when a person performs maximum muscle contraction is stored as the normalization parameter for the amplitude feature of each muscle. Alternatively, the average amplitude feature over a specified period, for example, up to 30 seconds after the start of a movement in which a person actually holds an object, may be used as the normalization parameter, or the normalization parameter may be determined by another method.

[0026] (Relative Muscle Activity Calculation Unit 40) The relative muscle activity calculation unit 40 calculates the normalized amplitude feature n k (t) is input, and the relative muscle activity r k (t) is calculated using the following formula (S3).

[0027]

[0028] The relative muscle activity calculation unit 40 calculates the relative muscle activity r k (t) is output to the muscle activity strength change calculation unit 50.

[0029] (Muscle activity strength change calculation unit 50) The muscle activity strength change calculation unit 50 calculates the time-series relative muscle activity r k Enter (t).

[0030] To obtain the reference muscle activity of each muscle, the muscle activity strength change calculation unit 50 calculates the average μ of the relative muscle activity of each muscle in a predetermined reference section. k and standard deviation σ k respectively.

[0031] The muscle activity strength change calculation unit 50 compares the muscle activity of each muscle with the muscle activity of the above-mentioned reference interval at each time, and labels the muscle activity strength in three levels: "strong," "medium," and "weak."

[0032] The strength label l of muscle activity at time t for the kth muscle k (t) is assigned as "-1," "1," or "0" as follows: The muscle activity strength label "-1" means the above-mentioned "weak," the muscle activity strength label "1" means the above-mentioned "strong," and the muscle activity strength label "0" means the above-mentioned "neutral."

[0033]

[0034] The muscle activity strength change calculation unit 50 calculates the following label matrix L obtained by labeling the strength of muscle activity for all muscles and times: L = [l k (t)] (k ∈ 1, ..., M, t ∈ 1, ..., T) In this way, the strength change of muscle activity is calculated (S4). The muscle activity strength change calculation unit 50 outputs the label matrix L to the time-series strength distribution calculation unit 60.

[0035] (Time-series strength distribution calculation unit 60) The time-series strength distribution calculation unit 60 receives the label matrix L from the muscle activity strength change calculation unit 50. The time-series strength distribution calculation unit 60 calculates the time proportion of each label in the label matrix L for each muscle.

[0036] For example, suppose that the strength label "-1" for muscle activity of muscle k occurs intermittently for 30 seconds, the strength label "0" for muscle activity of muscle k occurs intermittently for 60 seconds, and the strength label "1" for muscle activity of muscle k occurs intermittently for 10 seconds. In this case, the time percentage, which is the proportion of time the strength label "-1" occurred relative to the total time the labels "-1," "0," and "1" occurred, is calculated as 30 (= 30 / (30 + 60 + 10))%. Similarly, the time percentage for the strength label "0" is calculated as 60%, and the time percentage for the strength label "1" is calculated as 10%.

[0037] That is, the time series strength distribution vector s k is expressed as follows: sk = [time proportion of strength label "-1", time proportion of strength label "0", time proportion of strength label "1"] = [30, 60, 10]. The time series strength distribution calculation unit 60 calculates the time proportions for each muscle, and generates a time series strength distribution matrix S (= [s k ] (k=1, ..., M) (S5). The time-series intensity distribution calculation unit 60 outputs the matrix S of this time-series intensity distribution to the intensity bias calculation unit 70.

[0038] (Strength bias calculation unit 70) The strength bias calculation unit 70 receives the time-series strength distribution matrix S from the time-series strength distribution calculation unit 60. The strength bias calculation unit 70 uses this time-series strength distribution matrix S to determine the strength bias of the muscle activity amount for each muscle (S6).

[0039] The strength bias calculation unit 70 calculates the strength bias c of the muscle activity of muscle k. k is calculated as follows: c k = max(s k )-min(s k ) Note that max(s k ) is the time series strength distribution vector s k is a function that finds the maximum value from k ) is the time series strength distribution vector s k For example, the time series strength distribution vector s k If s is [30, 60, 10], then max(s k ) and min(s k ) is the following value: max(s k )=60, min(s k )=10 In other words, the bias in the strength of muscle activity of muscle k is c k is calculated as follows: c k = max(s k )-min(s k ) = 60 - 10 = 50 A relatively large difference between the maximum and minimum values ​​indicates that the muscle activity of the muscle in question is constantly in one of the strength labels "strong," "normal," or "weak," and that the duration of other strength label states is significantly shorter.

[0040] On the other hand, the fact that the difference between the maximum and minimum values ​​is relatively small means that the strength labels "strong," "normal," and "weak" of the muscle activity of the relevant muscle are distributed relatively evenly over time, and can be used as an indicator of how a person adjusts the strength of muscle activity when holding an object continuously.

[0041] In the embodiment described above, by adding a method for analyzing time-series changes in strength to the conventional method for analyzing isometric exercise, it is possible to appropriately evaluate the skill of a person in an exercise in which they continue to hold an actual object.

[0042] In this embodiment, athletic skill evaluation may be performed based on an index for a specific muscle, i.e., strength bias, or athletic skill evaluation may be performed based on the number of muscles for which this index falls below a certain threshold.

[0043] 3 is a block diagram showing an example of the hardware configuration of a skill evaluation index calculation device according to one embodiment of the present invention. In the example shown in FIG. 3, the skill evaluation index calculation device 100 according to the embodiment is configured, for example, by a server computer or a personal computer, and has a hardware processor 111A such as a CPU. A program memory 111B, a data memory 112, an input / output interface 113, and a communication interface 114 are connected to this hardware processor 111A via a bus 115.

[0044] The communication interface 114 includes, for example, one or more wireless communication interface units, and enables transmission and reception of information to and from a communication network NW. As the wireless interface, for example, an interface that adopts a low-power wireless data communication standard such as a wireless LAN (Local Area Network) is used.

[0045] An input device 300 and an output device 400 used by a user or the like, which are attached to the skill evaluation index calculation device 100, are connected to the input / output interface 113. The input / output interface 113 takes in operation data input by a user or the like through the input device 300, such as a keyboard, a touch panel, a touchpad, or a mouse, and outputs output data to an output device 400, which includes a display device using a liquid crystal or an organic electroluminescence (EL) display, for display. Note that the input device 300 and the output device 400 may be devices built into the skill evaluation index calculation device 100, or may be input devices and output devices of other information terminals that can communicate with the skill evaluation index calculation device 100 via a network NW.

[0046] The program memory 111B is a non-transitory tangible storage medium that is a combination of a non-volatile memory that can be written to and read from at any time, such as a hard disk drive (HDD) or a solid state drive (SSD), and a non-volatile memory such as a read only memory (ROM), and stores programs necessary to execute various control processes, etc., according to one embodiment.

[0047] The data memory 112 is a tangible storage medium that is a combination of, for example, the above-mentioned nonvolatile memory and a volatile memory such as RAM (Random Access Memory), and is used to store various data acquired and created during various processing steps.

[0048] The skill evaluation index calculation device 100 according to one embodiment of the present invention may be configured as a data processing device having a software-based processing function unit. The storage area used as a work memory or the like by the skill evaluation index calculation device 100 may be configured by using the data memory 112 shown in Fig. 3. However, these configured storage areas are not essential components within the skill evaluation index calculation device 100, and may be areas provided in, for example, an external storage medium such as a USB (Universal Serial Bus) memory, or a storage device such as a database server located in the cloud.

[0049] The processing function unit can be realized by having the hardware processor 111A read and execute a program stored in the program memory 111B, but the processing function unit may also be realized in various other forms, including an integrated circuit such as an application specific integrated circuit (ASIC) or a field-programmable gate array (FPGA).

[0050] The methods described in each embodiment may be stored as a program (software means) that can be executed by a computer on a recording medium such as a magnetic disk (e.g., a floppy disk, a hard disk, etc.), an optical disk (e.g., a CD-ROM, a DVD, an MO, etc.), or a semiconductor memory (e.g., a ROM, a RAM, a flash memory, etc.), or may be transmitted and distributed via a communication medium. The program stored on the medium also includes a configuration program that configures the software means (including not only executable programs but also tables and data structures) that the computer executes. The computer that realizes this device reads the program stored on the recording medium and, in some cases, configures the software means using the configuration program, and executes the above-described processing by having the operation controlled by this software means. The term "recording medium" as used herein is not limited to a storage medium for distribution, but also includes a storage medium such as a magnetic disk or semiconductor memory installed inside the computer or in a device connected via a network.

[0051] The present invention is not limited to the above-described embodiments, and various modifications can be made in the implementation stage without departing from the spirit of the invention. Furthermore, the embodiments may be implemented in appropriate combinations, in which case the combined effects can be obtained. Furthermore, the above-described embodiments include various inventions, and various inventions can be extracted by combining selected elements from the disclosed elements. For example, if the problem can be solved and the desired effect can be obtained even if some elements are deleted from all elements shown in the embodiments, the configuration from which these elements are deleted can be extracted as an invention.

[0052] 100...Skill evaluation index calculation device 10...Surface electromyogram measurement unit 20...Normalization parameter storage unit 30...Normalization unit 40...Relative muscle activity calculation unit 50...Muscle activity strength change calculation unit 60...Time series strength distribution calculation unit 70...Strength bias calculation unit

Claims

1. An evaluation index calculation device comprising: a first calculation unit that calculates amplitude features of a time series of electromyograms of each muscle of a subject; a second calculation unit that calculates a relative muscle activity of each muscle at each time based on a normalized value of the amplitude features calculated by the first calculation unit; a third calculation unit that calculates information indicating the strength of the calculated relative muscle activity based on the relative muscle activity calculated by the second calculation unit and the relative muscle activity in a reference interval; a fourth calculation unit that calculates a distribution of times during which the strength and weakness are occurring, as indicated by the information indicating the strength and weakness of the relative muscle activity calculated by the third calculation unit; and a fifth calculation unit that calculates a bias in the activity of each muscle of the subject based on the distribution of times during which the strength and weakness are occurring calculated by the fourth calculation unit.

2. The evaluation index calculation device of claim 1, wherein the third calculation unit calculates an average and standard deviation of the relative muscle activity in the reference section, and when the relative muscle activity calculated by the second calculation unit exceeds the sum of the calculated average and standard deviation, calculates information indicating the strength of the muscle activity that exceeds the sum of the calculated average and standard deviation, indicating that the muscle activity that exceeds the sum is relatively strong, and when the relative muscle activity calculated by the second calculation unit falls below the difference between the calculated average and standard deviation, calculates information indicating the strength of the muscle activity that falls below the sum is relatively weak.

3. The evaluation index calculation device described in claim 1, wherein the second calculation unit normalizes the amplitude feature calculated by the first calculation unit using a normalization parameter which is an amplitude feature related to a predetermined muscle activity, and calculates the relative muscle activity of each of the muscles at each time based on the normalized value.

4. A method performed by an evaluation index calculation device, comprising: a first calculation unit of the evaluation index calculation device calculating amplitude features of a time-series electromyogram of each muscle of a subject; a second calculation unit of the evaluation index calculation device calculating a relative muscle activity of each muscle at each time point based on a normalized value of the amplitude feature calculated by the first calculation unit; a third calculation unit of the evaluation index calculation device calculating information indicating the strength of the calculated relative muscle activity based on the relative muscle activity calculated by the second calculation unit and the relative muscle activity in a reference interval; a fourth calculation unit of the evaluation index calculation device calculating a distribution of times during which the strength and weakness are occurring, indicated by the information indicating the strength and weakness of the relative muscle activity calculated by the third calculation unit; and a fifth calculation unit of the evaluation index calculation device calculating a bias in the activity of each muscle of the subject based on the distribution of times during which the strength and weakness are occurring calculated by the fourth calculation unit.

5. The evaluation index calculation method according to claim 4, wherein the third calculation unit calculates an average and standard deviation of the relative muscle activity in the reference section, and when the relative muscle activity calculated by the second calculation unit exceeds the sum of the calculated average and standard deviation, calculates information indicating the strength of the muscle activity that exceeds the sum of the calculated average and standard deviation, and when the relative muscle activity calculated by the second calculation unit falls below the difference between the calculated average and standard deviation, calculates information indicating the strength of the muscle activity that falls below the sum of the calculated average and standard deviation, 6. The evaluation index calculation method according to claim 4, wherein the second calculation unit normalizes the amplitude feature calculated by the first calculation unit using a normalization parameter which is an amplitude feature related to a predetermined muscle activity, and calculates the relative muscle activity of each of the muscles at each time based on the normalized value.

7. An evaluation index calculation processing program that causes a processor to function as each part of the evaluation index calculation device according to any one of claims 1 to 3.

Citation Information

Patent Citations

  • Muscle activity analysis device, method, and program

    JP2018015405A

  • Rehabilitation evaluation device, rehabilitation evaluation method, and program

    JP2018038521A

  • Paraspinal electromyography scanning

    US5058602A

  • Myoelectric potential processing device, myoelectric potential processing method and myoelectric potential processing program

    WO2020070949A1