Motor function assessment device, motor function assessment method, and motor function assessment program

The motor function determination device analyzes motion and physical information to accurately estimate daily activities, overcoming limitations of existing technologies and enhancing rehabilitation effectiveness through comprehensive motor function assessments.

JP7595921B2Active Publication Date: 2024-12-09HIROSHIMA UNIVERSITY
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
JP2020208941
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2020-12-17
Publication Date
2024-12-09
Estimated Expiration
2040-12-17

AI Technical Summary

Technical Problem

Current motor function diagnostic devices struggle to accurately evaluate activities of daily living (ADL) due to reliance on limited movements, making it difficult to assess the adequacy of rehabilitation efforts and hinder the creation of data-based systems for medical fees and nursing care certification.

Method used

A motor function determination device that captures and analyzes a subject's motion video, assigns reference motion and physical information values, and estimates motor function through a combination of machine learning techniques, including support vector machines and neural networks, to provide comprehensive motor function assessments.

Benefits of technology

Enables accurate estimation of daily activities and motor functions, allowing for objective evaluation by medical professionals, improving the effectiveness of rehabilitation and facilitating data-based systems for medical and nursing care.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007595921000002
    Figure 0007595921000002
  • Figure 0007595921000003
    Figure 0007595921000003
  • Figure 0007595921000004
    Figure 0007595921000004
Patent Text Reader

Abstract

To provide a motion function determination device that is simple and easy to use for medical workers such as doctors, nurses, physical therapists, and care givers in sites of medical treatment, care, and rehabilitation, and allows movements in daily life to be reliably estimated from subject's limited movements.SOLUTION: A motion function determination device includes: a motion data acquisition unit for acquiring a motion video by imaging a predetermined motion of a subject and acquiring motion data on the subject by extracting a motion of the subject from the motion video; a motion value imparting unit for imparting a reference motion value corresponding to the motion data and generating a subject reference motion value; a body information acquisition unit for acquiring predetermined body information on the subject; a body information value imparting unit for generating a subject reference body information value by imparting a reference body information value corresponding to the body information; and an estimation unit for estimating a motion function of the subject by combining the subject reference motion value imparted to the subject with the subject reference body information value, and outputting an estimation result.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to a motor function assessment device, a motor function assessment method, and a motor function assessment program, and in particular to a motor function assessment device and a method and program having a function of estimating motor function from a specific movement of a subject. [Background technology]

[0002] Although the decline in motor skills due to aging is inevitable, it is possible to prevent this decline through rehabilitation provided by various medical institutions. In this case, it is assumed that the subject (patient, etc.) receiving the training will need to make an effort, and it is also necessary to accurately understand the subject's case and provide training content appropriate to the case. By providing appropriate training and rehabilitation, the decline in motor function can be prevented, and the quality of life (QOL) of the elderly can be maintained.

[0003] Currently, the understanding of subjects' cases is mainly done by visual assessment by medical professionals, and individual evaluation scores are determined for the assessment of activities of daily living (ADL: Activities of Daily Living) defined in the Functional Independence Measure (FIM) based on visual assessment. Based on such scores, medical institutions are regulated for medical treatment fees, and individuals are regulated for nursing care certification.

[0004] From the subject's point of view, it is difficult to imagine that rehabilitation and other training contributes to improving daily life. From the medical professional's point of view, it is not easy to evaluate activities of daily living (ADL), and the evaluation of the effects of rehabilitation after its implementation is unclear. Furthermore, from the administrative point of view, it is an obstacle to the construction of a data-based medical remuneration and nursing care certification system.

[0005] Therefore, there has been a demand for a technology that objectively links the subject's movements, motor abilities, and the evaluation of activities of daily living (ADL).For example, a motor function diagnostic device has been proposed that includes a movement detection unit that detects the movements of multiple predetermined body parts from an image of the subject, and a judgment unit that judges the subject's motor function based on the detection results of the movements of each part, calculates the acceleration of the movement of each predetermined part of the subject, obtains characteristic parameters from the acceleration, and judges the motor function using a neural network (see Patent Document 1).

[0006] However, the technology represented by Patent Document 1 judges motor function based on limited movements. Therefore, it is difficult to estimate the appropriateness of activities of daily living (ADL) from the movement alone. Therefore, existing motor function diagnostic devices still do not meet the standard required in clinical practice. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] JP 2016-144598 A Summary of the Invention [Problem to be solved by the invention]

[0008] The present invention has been made in consideration of the above-mentioned points, and provides a motor function assessment device, method, and program that enable medical professionals such as doctors, nurses, physical therapists, and care workers in medical, nursing, and rehabilitation settings to accurately estimate a subject's daily living activities, etc. from limited movements. [Means for solving the problem]

[0009] That is, the motor function assessment device of the embodiment includes a motion data acquisition unit that captures a predetermined motion of a subject to acquire a motion video and extracts the subject's motion from the motion video to acquire motion data of the subject, a motion value assignment unit that assigns a reference motion value corresponding to the motion data to generate a subject reference motion value, a physical information acquisition unit that acquires predetermined physical information of the subject, a physical information value assignment unit that assigns a reference physical information value corresponding to the physical information to generate a subject reference physical information value, and an estimation unit that estimates the motor function of the subject by combining the subject reference motion value assigned to the subject with the subject reference physical information value and outputs the estimation result.

[0010] Furthermore, the reference motion values ​​may be a plurality of numerical values ​​corresponding to a plurality of predetermined motions of the subject, respectively, and the reference physical information values ​​may be a plurality of numerical values ​​corresponding to a plurality of predetermined physical information of the subject, respectively.

[0011] Furthermore, the apparatus may include a reference motion value storage unit which stores in advance, prior to acquiring motion data of the subject, a plurality of reference motions of the subject associated with a plurality of reference motion values ​​corresponding to each of the plurality of reference motions, and a reference physical information value storage unit which stores in advance, prior to acquiring specified physical information of the subject, a plurality of physical information of the subject associated with a plurality of reference physical information values ​​corresponding to each of the plurality of physical information.

[0012] Furthermore, the motion image may be taken while the subject is seated.

[0013] Furthermore, the motor function may be the walking function of the subject.

[0014] Furthermore, the physical information may be selected from myoelectric potential, heart rate, and time spent maintaining a posture.

[0015] Furthermore, machine learning may be used to combine the subject's reference motion value and the subject's reference physical information value assigned to the subject, thereby estimating the subject's motor function. Effect of the Invention

[0016] According to the motor function assessment device of the present invention, the device is equipped with a motion data acquisition unit that captures specified movements of a subject to acquire motion footage and extracts the subject's movements from the motion footage to acquire motion data of the subject, a motion value assignment unit that assigns a reference motion value corresponding to the motion data to generate a subject reference motion value, a physical information acquisition unit that acquires specified physical information of the subject, a physical information value assignment unit that assigns a reference physical information value corresponding to the physical information to generate a subject reference physical information value, and an estimation unit that combines the subject reference motion value assigned to the subject with the subject reference physical information value to estimate the subject's motor function and output the estimation result, so that in medical, nursing care, and rehabilitation settings, medical professionals such as doctors, nurses, physiotherapists, and care workers can accurately estimate daily living activities, etc. from the subject's limited movements. [Brief description of the drawings]

[0017] [Figure 1] 1 is a schematic diagram showing a configuration of a motor function determination device according to an embodiment; [Diagram 2] 2 is a block diagram showing a configuration of a functional section of the motor function determination device. FIG. [Diagram 3] FIG. 1 is a schematic diagram relating to acquisition of reference motion and motion data. [Figure 4] FIG. 13 is a schematic diagram showing generation of a subject reference physical information value. [Diagram 5] FIG. 1 is a first schematic diagram showing a flow of estimating a subject's motor function. [Figure 6] FIG. 2 is a second schematic diagram showing the flow of estimating the motor function of a subject. [Figure 7] FIG. 1 is a schematic diagram showing a flow of estimating a subject's motor function using machine learning. [Figure 8] FIG. 1A is a schematic diagram showing an example of indirect torque, and FIG. 1B is a schematic diagram showing an example of a musculoskeletal model. [Figure 9] 1 is a first flowchart illustrating a motor function assessment method according to an embodiment. [Figure 10] 11 is a second flowchart illustrating the motor function assessment method according to the embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0018] The motor function assessment device of the embodiment is a device for estimating the level of a subject's (patient's) motor function from the analysis of limited movements of the subject in actual clinical settings of medical care, nursing care, and rehabilitation. By using a computer, objective judgment is realized in estimating the motor function of a subject (patient) from the empirical judgment made by a medical professional such as a doctor by visual inspection at present.

[0019] FIG. 1 is a schematic diagram showing the configuration of a motor function assessment device 1 according to an embodiment. The motor function assessment device 1 is composed of an imaging unit 3 that images a subject and a target person (described later), a processing unit 2 that analyzes the captured image and performs calculations to estimate motor function, and a display unit 4 that is connected to the processing unit 2 and performs various displays. Reference symbol H in the figure indicates a subject, who is sitting on a chair (seated state) with his right arm raised. Subject H is fitted with a heart rate monitor 21 and an electromyograph 22. The heart rate monitor 21 and the electromyograph 22 are equipped with radio wave transmitters (not shown) that emit radio waves compatible with wireless standards such as IEEE802.11ac and Bluetooth (registered trademark), and are capable of transmitting to the processing unit 2.

[0020] The image capturing unit 3 is a camera capable of capturing video, and may be a terminal such as a smartphone or tablet. The display unit 4 is a known display. The processing unit 2 is a personal computer (PC), a mainframe, a workstation, a cloud computing system, or various electronic computers (computing resources, computers) such as a smartphone or tablet. When the processing unit 2 is a smartphone or tablet, the processing unit 2 may be configured to include the image capturing unit 3 and the display unit 4.

[0021] In the embodiment, the subject is a person who receives nursing care, rehabilitation, etc., and is a subject of judgment (diagnosis) by a doctor (medical professional) in a clinical setting. The subject is a person whose motor function is estimated by the motor function determination device 1.

[0022] The subject is a person who performs a movement (reference movement) that is used as a standard for estimating the subject's motor function by the motor function assessment device 1, and may be a healthy person or a diseased person. In addition, the subject himself may be the subject.

[0023] The processing unit 2 is configured with hardware such as a microcomputer necessary for controlling various operations such as receiving signals from the imaging unit 3, executing calculations, storing data, etc., and is equipped with a CPU 10, ROM 11, RAM 12, I / O 13 (input / output interface), a memory unit 14, etc.

[0024] When each functional unit of the processing unit 2 (computer) in Fig. 1 is realized by software, the processing unit 2 is realized by executing instructions of a program, which is software that realizes each function. The recording medium that stores this program can be a "non-transitory tangible medium," such as a CD, a DVD, a semiconductor memory, or a programmable logic circuit. In addition, this program may be supplied to the processing unit 2 of the motor function assessment device 1 via any transmission medium (communication network, broadcast wave, etc.) that can transmit the program.

[0025] The memory unit 14 of the processing unit 2 is a known storage device such as an HDD or SSD. The memory unit 14 may be an external server (not shown). The memory unit 14 stores various data, information, a motor function assessment program, various data required for executing the program, and the like. Each functional unit that executes various calculations, arithmetic operations, and the like is a computing element such as a CPU 10. In addition, input devices (not shown) such as a keyboard and a mouse, a display unit 4 (a display device such as a display), an output device that outputs data, and the like may also be appropriately connected to the I / O 13 of the processing unit 2.

[0026] The functional units in the CPU 10 of the processing unit 2 are shown in the schematic block diagram of Fig. 2. The functional units include a reference motion value storage unit 110, a reference physical information value storage unit 120, a motion data acquisition unit 210, a motion value assignment unit 220, a physical information acquisition unit 230, a physical information value assignment unit 240, and an estimation unit 250. The operation and execution of the processing unit 2 is realized in software terms by a motor function assessment program loaded into the main memory, etc.

[0027] The reference motion value storage unit 110 stores a plurality of reference motions of the subject in association with a plurality of reference motion values ​​corresponding to each of the plurality of reference motions. Examples of reference motions include bending and straightening the knees and elbows, clenching and opening the hands, tilting the upper body back and forth and side to side, swinging the arms, turning the head, and even various walking movements (without a cane and with a cane). The numerical value of the reference motion value is set appropriately. When the subject is completely unable to perform the reference motion, it can be set to "0", and depending on the degree of the difficulty, it can be set to a value on a 5-level scale (highest is "5") or 10-level scale (highest is "10").

[0028] The function of the reference motion value storage unit 110 will be described with reference to Fig. 3. In the motor function determination device 1, first, a plurality of reference motions required for estimating motor function are photographed through the imaging unit 3. In the illustrated example, the reference motions are performed by the subject. In Fig. 3, the subject is photographed while sitting in a chair. Although not illustrated, it is also possible to photograph the subject in a standing position. Starting from the top in the figure, the subject is asked to move the upper limbs (angle of bending and straightening the elbow), fingers, hip joint (angle of opening), knee joint (bending and straightening the knee), and ankle joint as a plurality of reference motions 30 on the left and right sides of the body, and photographed through the imaging unit 3.

[0029] The multiple reference movements captured by the imaging unit 3 are converted into skeletal movements in the processing unit 2. In the illustrated example, the multiple reference movements are displayed as a rod-shaped model 31 indicating joint positions displayed next to the captured image. Additionally, multiple reference movement values ​​32 are displayed next to the rod-shaped model 31. The illustration is merely an example of the reference movements, and the reference movement values ​​(specific numerical values) are defined according to the degree of each individual reference movement.

[0030] For example, the reference motion value is 5 points when all five fingers are curled up to make a fist, 4 points if only partway through, or 3 points depending on the stage. In the case of the hip joint, 5 points are given if the legs are spread 60 degrees from the state where both knees are together, and 4 points are given if the legs are spread 45 degrees. Of course, the reference motion value can be determined by taking into account the time required for the motion (elapsed time). In this way, a plurality of reference motion values ​​are linked in correspondence with a plurality of reference motions like a quick reference table in the form of a table. A plurality of reference motions and a plurality of corresponding reference motion values ​​are stored in the memory unit 14 of the processing unit 2.

[0031] The reference physical information value storage section 120 stores a plurality of pieces of physical information of the subject and a plurality of reference physical information values ​​corresponding to each of the plurality of pieces of physical information in association with each other.

[0032] The function of the reference physical information value storage unit 120 will be described with reference to Fig. 4. In this embodiment, the multiple pieces of physical information of the subject are information that can be objectively grasped from the outside and quantified, such as the myoelectric potential of muscles in various parts of the body, heart rate, and even the time during which the same posture can be maintained. Therefore, for the physical information acquired from the subject, a range of the physical information is defined, and a reference physical information value is assigned as a numerical value (index) corresponding to the range of the physical information. The multiple pieces of physical information and the multiple reference physical information values ​​corresponding to them are stored in the memory unit 14 of the processing unit 2.

[0033] For example, in FIG. 4(A), the range of values ​​of myoelectric potential (physical information) in the muscles of the subject's thighs, upper arms, etc. are associated with scores (reference physical information values) corresponding to the range of values. In FIG. 4(B), the range of values ​​of heart rate (physical information) are associated with scores (reference physical information values) corresponding to the range of values. In FIG. 4(C), the range of values ​​of the time (physical information) for maintaining a posture with knees stretched straight out in front while seated is associated with scores (reference physical information values) corresponding to the range of values. Multiple reference physical information values ​​are linked in a quick reference table format with vertical and horizontal axes corresponding to multiple pieces of physical information.

[0034] When acquiring myoelectric potential as physical information, both the subject and the test subject are fitted with a myoelectric potential detection sensor. When acquiring heart rate, both the subject and the test subject are fitted with a pulse meter. When acquiring posture maintenance time, both the subject and the test subject are analyzed from the video captured by the imaging unit 3.

[0035] In this way, prior to estimating the motor function of a subject, the subject is asked to perform a plurality of reference movements, and a plurality of pieces of physical information are obtained from the subject. Then, a reference movement value and a reference physical information value that serve as the basis for evaluation and classification are established in advance. When establishing the reference movement value and the reference physical information value, gender differences, age differences, regional differences, and the like are also taken into consideration. The reference movement value and the reference physical information value may be established once, or may be updated as necessary.

[0036] Next, the motion data acquiring section 210 captures a predetermined motion of the subject to acquire a motion video, and extracts the motion of the subject from the motion video to acquire motion data of the subject.

[0037] The subject is photographed in the same manner as in the photographing of the reference movement described in FIG. 3 above. As shown in FIG. 3, the subject is asked to perform a specified movement, and the state of the movement is photographed as a movement image through the photographing unit 3. As with the photographing of the target person, the subject is photographed while sitting in a chair. Although not shown, it is also possible to photograph the subject in a standing position. Since the specific movements are common, their explanation is omitted. The process of extracting the subject's movement from the movement image is to create a rod-shaped model 31 that replaces the movement of the subject's skeleton, as shown in FIG. 3. Since the movement data is the drive amount of the joints (the range that can be moved), the elements for analyzing the movement of the joints of the subject's body are reduced, making it easier to understand the subject's movement.

[0038] The motion value assigning unit 220 assigns a reference motion value corresponding to the motion data to generate a subject's reference motion value. When assigning a reference motion value, as shown in Fig. 3, a reference motion value 32 for each motion corresponding to the motion data (individual motion) that has become a rod-shaped model 31 is displayed. The assigned reference motion value 32 is then generated as the subject's reference motion value. The subject's reference motion value is stored in the memory unit 14 of the processing unit 2.

[0039] The physical information acquiring section 230 acquires predetermined physical information of the subject.

[0040] The acquisition of the predetermined physical information of the subject is similar to the case of the subject described above, and involves acquiring information that can be objectively grasped from the outside and quantified about the subject's body, such as the myoelectric potential of muscles in various parts of the body, heart rate, the time the same posture can be maintained, etc. One or more of the myoelectric potential, heart rate, and the time the same posture can be maintained may be selected.

[0041] The physical information value assigning section 240 assigns a reference physical information value corresponding to the physical information to generate a subject reference physical information value.

[0042] Even in the case of a subject, as shown in FIG. 4, the range of the subject's myoelectric potential (physical information) in FIG. 4(A) is associated with a score (subject's reference physical information value) corresponding to the range of the value. The range of the subject's heart rate (physical information) in FIG. 4(B) is associated with a score (subject's reference physical information value) corresponding to the range of the value. The range of the subject's posture maintenance time (physical information) in FIG. 4(C) is associated with a score (subject's reference physical information value) corresponding to the range of the value. A plurality of subject's reference physical information values ​​are linked in a quick reference table in a table format with vertical and horizontal axes corresponding to a plurality of pieces of physical information. Note that the type of physical information that is the basis of the subject's reference physical information value in FIG. 4 is an example, and physical information other than that shown and described may be included.

[0043] The estimation unit 250 estimates the motor function of the subject by combining the subject's reference motion value and the subject's reference physical information value. The estimated motor function is output as an estimation result. The estimation result is output through the display unit 4.

[0044] An overview of the processing of the estimation unit 250 is shown in FIG. 5. The motor function of the subject is estimated by considering two types of information: the subject's reference motion value obtained by having the subject perform a predetermined motion, and the subject's reference physical information obtained from the subject's physical information as information other than the motion. When estimating the motor function, the types of combinations of the subject's reference motion value and the subject's reference physical information are predefined, and are linked like a quick reference table in a table format with vertical and horizontal axes. For example, the combination of the subject's reference motion value "3" and the subject's reference physical information "4" is linked to the estimation result of the motor function "Independent walking is difficult."

[0045] A subject's reference motion value is generated based on a plurality of reference motion values ​​and the actual motion of the subject. Also, a subject's reference information value is generated based on a plurality of reference physical information values ​​and the actual physical information of the subject. Then, a target motor function is estimated from the numerical values ​​of both the subject's reference motion value and the subject's reference information value.

[0046] The most representative motor function referred to here is the ability to walk. Being able to walk improves QOL (quality of life). Of course, in addition to walking function, various other motor functions are also included, such as being able to eat alone, buttoning one's clothes alone, and disposing of excretions alone. More specifically, motor function is estimated to be evaluated according to the Activities of Daily Living (ADL) defined in the Functional Independence Measure (FIM). Of these, 13 motor ADL items can be used to evaluate motor function (see Table 1).

[0047] [Table 1]

[0048] Basic information for estimation can be obtained from the subject's reference motion value generated by associating the reference motion and the corresponding reference motion value disclosed in FIG. 3. However, the subject's reference motion value only obtains information on one aspect of the motion. For this reason, the accuracy of the estimation tends to be low. Therefore, the accuracy of the estimation can be improved by adding the subject's reference body information that takes into account the load on the body during the motion as well as the strain on the muscles and heart, in addition to the subject's reference motion value.

[0049] For example, as shown in the schematic diagram of FIG. 6, if the subject's reference motion value was "4" and the subject's reference information value was "5" in the previous measurement, and the subject's reference motion value is "4" and the subject's reference information value is "6" this time (for example, 4 weeks later), there is no change in the subject's reference motion value. In this case, estimation based only on the subject's reference motion value can only result in no improvement in motor function. However, in reality, the subject's reference information value is increasing, and a certain improvement is seen in heart rate and muscles.

[0050] Therefore, it can be estimated that the actual motor function of the subject will improve if the subject continues rehabilitation for another three weeks. As a result, it becomes possible to grasp the motor function of each subject in detail. This can then serve as an opportunity to modify or change the content of rehabilitation depending on the estimation results. It also makes it possible to give meaning to the subject's decision to continue rehabilitation. Furthermore, it makes it possible to objectively show changes in the subject's rehabilitation.

[0051] In addition, in the case of a subject who has difficulty walking (exclusively using a wheelchair), it is self-evident that it is not possible to obtain motion data in a standing position, and only motion data in a sitting position can be obtained. However, although no change is observed in the subject's reference motion value, functional improvement can be read from the subject's reference physical information by continuing rehabilitation, etc. In this case, even if walking is currently difficult, the subject can be informed of the estimated result that walking will be possible in the future, thereby increasing the subject's motivation for rehabilitation, etc. In this way, the motor function assessment device 1 is not necessarily limited to only the motions that the subject is currently capable of, but can realize comprehensive judgment and estimation including physical information, and can provide feedback of the estimated results to the subject, medical professionals, and government officials.

[0052] The estimation unit 250 may further use machine learning to combine the subject's reference motion value and the subject's reference physical information value assigned to the subject, thereby estimating the motor function of the subject. There are a wide variety of reference motion values ​​prior to the combination of the subject's reference motion value and the subject's reference physical information value. Therefore, machine learning techniques such as Support Vector Machine (SVM), model tree, decision tree, neural network, multiple linear regression, locally weighted regression, and probabilistic search method are used for estimation using the subject's reference motion value and the subject's reference physical information value.

[0053] Fig. 7 is a schematic diagram of a case where a neural network is used, for example. The values ​​of "heart rate variability, joint position 1, joint position 2, joint angle 1, joint angle 2, joint torque 1, joint torque 2, muscle force 1, muscle force 2, ..." of a certain subject are input as an Input Layer. In this example, "heart rate variability" is the subject's reference physical information value or reference physical information value, and the rest are the subject's reference motion values ​​or reference motion values.

[0054] 7, for "joint position 1, joint position 2, joint angle 1, joint angle 2," the joint positions and angles are estimated from motion information using a known bone model. Furthermore, for "joint torque 1, joint torque 2, muscle force 1, muscle force 2," dynamic information (joint torque, muscle force) is estimated using a known musculoskeletal model.

[0055] For example, the correlation between a plurality of reference movement values ​​and reference physical information values ​​and the previously known motor function of the subject (ADL score in the figure) is machine-learned. Then, a model for statistical estimation of motor function is formed from the reference movement values ​​and reference physical information values ​​through the results of machine learning. Then, the subject's reference movement values ​​and subject's reference physical information values ​​are inputted into the model for estimating motor function from the reference movement values ​​and reference physical information values, and the motor function (ADL score) of the subject is estimated. Finally, the ADL score (see Table 1) is presented on the Output Layer as the estimation result of motor function. The estimation result is output through the display unit 4.

[0056] The schematic diagram of FIG. 8 is an example of a known analysis method used to obtain the reference motion value and the subject's reference motion value. FIG. 8(A) is a schematic diagram of ergonomics evaluation, in which the subject and the subject walking are photographed and motion data is generated focusing on the motion of the joints. This process is used to estimate and evaluate the joint torque of the human body. FIG. 8(B) is a schematic diagram of a whole-body musculoskeletal model omitting both arms. The subject and the subject standing are photographed and motion data is generated focusing mainly on the motion of the skeleton and muscles of the lower body. Of course, various methods for analyzing the joints, muscles, and skeleton of the body are used in addition to those shown in the figure.

[0057] 9 and 10, the motor function assessment method and the motor function assessment walking training support program in the motor function assessment device 1 will now be described. The motor function assessment method is executed by the CPU 10 of the processing unit 2 based on the motor function assessment program. The motor function assessment program causes the processing unit 2 (computer) in FIG. 1 to execute various functions, including a reference motion value storage function, a reference physical information value storage function, a motion data acquisition function, a motion value assignment function, a physical information acquisition function, a physical information value assignment function, and an estimation function. Each of these functions is executed in the order shown in the figure. Note that each function overlaps with the description of the motor function assessment device 1 described above, so details will be omitted.

[0058] As can be seen from the flowchart of Fig. 9, the processing by the processing unit 2 includes various steps such as a reference motion value retaining step (S110) and a reference physical information value retaining step (S110). Furthermore, as can be seen from the flowchart of Fig. 10, the processing by the processing unit 2 includes various steps such as a motion data acquiring step (S210), a motion value assigning step (S220), a physical information acquiring step (S230), a physical information value assigning step (S240), and an estimation step (S250). Of course, various steps necessary for the operation of the processing unit 2 itself are naturally included.

[0059] The reference motion value storage function stores a plurality of reference motions of the subject in association with a plurality of reference motion values ​​corresponding to each of the plurality of reference motions (S110; reference motion value storage step). The reference physical information value storage function stores a plurality of physical information of the subject in association with a plurality of reference physical information values ​​corresponding to each of the plurality of physical information (S120; reference physical information value storage step). The reference motion value storage function and the reference physical information value storage function are executed by the CPU 10 of the processing unit 2.

[0060] The motion data acquisition function captures a predetermined motion of the subject to acquire a motion image, and extracts the motion of the subject from the motion image to acquire motion data of the subject (S210; motion data acquisition step). The motion value assignment function assigns a reference motion value corresponding to the motion data to generate a subject reference motion value (S220; motion value assignment step). The physical information acquisition function acquires predetermined physical information of the subject (S230; physical information acquisition step). The physical information value assignment function assigns a reference physical information value corresponding to the physical information to generate a subject reference physical information value (S240; physical information value assignment step). The estimation function combines the subject reference motion value assigned to the subject with the subject reference physical information value to estimate the motor function of the subject and output the estimation result (S250; estimation step). Each of these functions is executed by the CPU 10 of the processing unit 2.

[0061] The computer program of the present invention described above may be recorded on a processor-readable recording medium, and the recording medium may be a "non-transitory tangible medium" such as a tape, disk, card, semiconductor memory, programmable logic circuit, etc.

[0062] The computer program can be implemented using, for example, a scripting language such as ActionScript or JavaScript (registered trademark), an object-oriented programming language such as Objective-C or Java (registered trademark), or a markup language such as HTML5. [Industrial Applicability]

[0063] According to the motor function assessment device of the present invention, it is possible to efficiently assess motor function by simply preparing a simple device, and it is also possible to ensure objectivity in the assessment. Therefore, it is promising as equipment for assessing motor function in clinical practice. [Explanation of symbols]

[0064] 1 Motor function determination device 2 Processing section 3. Photography Department 4 Display section 10 CPU 11 ROM 12 RAM 13 I / O 14 Storage section 21 Heart Rate Monitor 22 Electromyograph 30 Reference Operation 31 Rod Model 32 Reference operating value 110 Reference operating value storage unit 120 Reference body information value storage unit 210 Operation data acquisition unit 220 Operation value assignment unit 230 Physical Information Acquisition Department 240 Physical information value assignment unit 250 Estimation section H. Subject

Claims

1. A reference motion value storage unit that stores in advance a plurality of reference motions of a subject, the subject being asked to perform motions that are to be used as a reference for estimating the subject's motor function, and that associates a plurality of reference motion values, which are a plurality of numerical values ​​corresponding to a plurality of predetermined motions of the subject, with the plurality of reference motions, respectively, and stores the plurality of reference motion values ​​in advance, the plurality of numerical values ​​corresponding to the plurality of predetermined motions of the subject. a reference physical information value storage unit that stores in advance a plurality of reference physical information values, each of which corresponds to a plurality of predetermined physical information of the subject, in association with each other, the plurality of reference physical information values ​​being a plurality of numerical values ​​corresponding to each of the plurality of predetermined physical information of the subject, the plurality of reference physical information values ​​being corresponding to each of the plurality of predetermined physical information of the subject, a motion data acquisition unit that acquires a motion image by photographing a predetermined motion of the subject, and extracts the motion of the subject from the motion image to acquire motion data of the subject; a motion value assigning unit that assigns the reference motion value corresponding to the motion data to generate a subject reference motion value; a physical information acquiring unit for acquiring predetermined physical information of the subject; a physical information value assigning unit that assigns the reference physical information value corresponding to the physical information to generate a subject reference physical information value; an estimation unit that estimates a motor function of the subject by combining the subject's reference motion value and the subject's reference physical information value, and outputs an estimation result; Equipped A motor function assessment device comprising:

2. 2. The motor function determination device according to claim 1, wherein the motion image is taken while the subject is in a seated position.

3. The motor function assessment device according to claim 1 or 2, wherein the motor function is the walking function of the subject.

4. 4. The motor function determination device according to claim 1, wherein the physical information is selected from a myoelectric potential, a heart rate, and a time period during which a posture is maintained.

5. The motor function determination device according to any one of claims 1 to 4, wherein machine learning is used to combine the subject's reference motion value and the subject's reference physical information value assigned to the subject, thereby estimating the subject's motor function.

6. The computer a reference motion value storage step in which, prior to acquiring motion data of a subject, a plurality of reference motions of the subject who is to perform motions that are to be used as a reference for estimating the motor function of the subject are photographed, and a plurality of reference motion values, which are a plurality of numerical values ​​corresponding to each of a plurality of predetermined motions of the subject, are associated with each of the plurality of reference motions and stored in advance; a reference physical information value storage step of acquiring a plurality of pieces of physical information of the subject prior to acquiring the predetermined physical information of the subject, and associating a plurality of reference physical information values, which are a plurality of numerical values ​​corresponding to each of the plurality of pieces of physical information of the subject, with each of the plurality of pieces of physical information and each of the plurality of predetermined physical information of the subject, and storing the reference physical information value in advance; a motion data acquisition step of photographing a predetermined motion of the subject to acquire a motion image, and extracting the motion of the subject from the motion image to acquire motion data of the subject; a motion value assigning step of generating a subject reference motion value by assigning the reference motion value corresponding to the motion data; a physical information acquiring step of acquiring predetermined physical information of the subject; a physical information value assigning step of generating a subject reference physical information value by assigning the reference physical information value corresponding to the physical information; an estimation step of estimating the motor function of the subject by combining the subject's reference motion value assigned to the subject with the subject's reference physical information value and outputting an estimation result; A method for assessing motor function.

7. The computer a reference motion value storage unit that stores in advance a plurality of reference motions of a subject, the subject being asked to perform a motion that is to be used as a reference for estimating the motor function of the subject, in which the plurality of reference motions are captured prior to acquisition of motion data of the subject, and a plurality of reference motion values, which are a plurality of numerical values ​​corresponding to a plurality of predetermined motions of the subject, are associated with the plurality of reference motions, respectively; a reference physical information value storage unit that stores in advance a plurality of reference physical information values, which are a plurality of numerical values ​​corresponding to each of the plurality of predetermined physical information of the subject, associated with each of the plurality of pieces of physical information of the subject, and stores the plurality of reference physical information values ​​in advance, prior to acquiring the predetermined physical information of the subject; a motion data acquisition function of acquiring a motion image by photographing a predetermined motion of the subject, and acquiring motion data of the subject by extracting the motion of the subject from the motion image; a motion value assigning function for generating a subject reference motion value by assigning the reference motion value corresponding to the motion data; a physical information acquisition function for acquiring predetermined physical information of the subject; a physical information value assigning function for generating a subject reference physical information value by assigning the reference physical information value corresponding to the physical information; and an estimation function of estimating the motor function of the subject by combining the subject's reference motion value given to the subject with the subject's reference physical information value and outputting the estimation result. A motor function assessment program comprising:

Citation Information

Patent Citations

  • Motor performance detecting device

    JP2009101108A

  • Motor function diagnostic device, method, and program

    JP2016144598A

  • Motor function estimation information generating device, motor function estimating system, motor function estimation information generating method, motor function estimating method, and recording medium

    JP2018033949A

  • Motion analyzing device, motion analyzing method, and motion analyzing program

    WO2018047847A1