JP6917047B2 - Motor function evaluation system with multiple wearable sensors - Google Patents

Description

The present invention relates to a system for evaluating the risk of locomotive syndrome as a subject's motor function by a plurality of wearable sensors.

Traditionally, the term locomotive syndrome has been proposed by the Japanese Orthopedic Association. "Locomotive Syndrome" (also abbreviated as "locomotive syndrome") is a term that means "a high risk of needing care due to a disorder of the locomotive syndrome" (Japanese clinical practice). Quoted from the Orthopedic Association HP: See Non-Patent Document 1). Assessing the risk of this locomotive syndrome is also important to avoid becoming in need of nursing care in the future.

Here, Patent Document 1 discloses a technique for evaluating the risk of senile disability. Further, Patent Document 2 discloses a motor function diagnostic device for evaluating the risk of locomotive syndrome.

JP 2013-255786 JP 2016-144598

"Locomotive Syndrome", [online], March 3, 2009, Japanese Society of Clinical Orthopedic Surgery, [Search November 28, 2016], Internet <URL: http://www.jcoa.gr.jp/ locomo / teigi.html>

However, the risk of senile disability, which is the subject of evaluation in Patent Document 1, does not include locomotive syndrome. Further, the evaluation method in Patent Document 1 uses walking parameters.

The motor function diagnostic apparatus disclosed in Patent Document 2 is intended for risk assessment of locomotive syndrome. However, the Kinect (registered trademark) sensor is used to detect the movement of multiple parts of the subject's body, and such a system requires evaluation at a specific location where the sensor is installed, and during daily life. Cannot be evaluated. In addition, there is a problem that the detection cannot be performed when the subject is wearing a skirt or the like and the part to be detected is hidden.

In view of these problems, the present invention uses a plurality of wearable sensors to more accurately evaluate the risk of locomotive syndrome as a motor function and provide feedback to the user even if the sensor is not installed in a specific place. It is an object of the present invention to provide a motor function evaluation system capable of performing.

In order to solve the above problems, one aspect of the present invention is a motor function evaluation system that evaluates the risk of locomotive syndrome as a subject's motor function and feeds back the result to the user, and is a motor function evaluation system on the subject's body. Data for acquiring a plurality of wearable sensor units attached to a plurality of specific parts and a plurality of measurement data indicating changes in the plurality of specific parts of the body of the subject in a walking motion from the plurality of sensor units. The locomotive degree score estimation unit that estimates the locomotive risk of the subject by calculating the variation between the acquisition unit and the plurality of measurement data or the data derived from the plurality of measurement data, and the estimation of the locomotive risk of the subject. It is a motor function evaluation system having a result output unit that outputs a result.

In addition, another aspect of the present invention includes a plurality of wearable sensor units, a data acquisition unit, a locomotive degree score estimation unit, and a result output unit, which are attached to a plurality of specific parts on the body of the subject. It is a motor function evaluation method executed by a computer system having the data acquisition unit, and the data acquisition unit obtains a plurality of measurement data indicating changes in a plurality of specific parts of the body of the subject in a walking motion by the plurality of sensor units. The step of estimating the locomotive risk of the subject by calculating the variation of the plurality of measurement data or the data derived from the plurality of measurement data by the locomotive degree score estimation unit, and the result. The output unit is a motor function evaluation method including a step of outputting the estimation result of the locomotive risk of the subject.

Another aspect of the present invention is a computer program for causing the computer system to execute the above-mentioned motor function evaluation method.

It is a figure which shows an example of the functional block of the motor function evaluation system which concerns on one Embodiment of this invention. It is a figure which shows an example of the hardware structure of the motor function evaluation system (motor function evaluation apparatus) which concerns on one Embodiment of this invention. It is a figure which shows an example of the flowchart about the preparatory stage for performing the locomotive risk evaluation which concerns on one Embodiment of this invention. It is a figure explaining an example of construction of the discrimination model in the locomotive risk evaluation method which concerns on one Embodiment of this invention. It is a figure explaining an example of a discrimination model. It is a figure which shows an example of the flowchart for performing locomotive risk evaluation in the motor function evaluation system which concerns on one Embodiment of this invention.

(overview)
Prior to the invention of the motor function evaluation system for evaluating the risk of locomotive syndrome (hereinafter, abbreviated as "locomotive syndrome") according to the present invention, the inventor of the present invention has a higher locomotive risk than a low locomotive risk. When walking, there is a large variation in movement from the late stance phase (standing phase: the state where the foot is in contact with the ground during walking) to the early stage of swinging (the swing phase: the state where the foot is off the ground during walking). (Yoshiyuki Kobayashi, 3 others, "Gait Characteristics of high scorers of the locomotive syndrome", Journal of the Society of Biomechanisms Japan), Society of Biomechanism, Vol.40, No.3, 2016, P183-193). As a result, in the motor function evaluation system according to the present invention, the locomotive risk of the subject is evaluated based on the variation in the measurement data of the walking motion of the subject.

In addition, in the motor function evaluation system according to the present invention, various measurement data are collected in advance from Locomo applicable persons and Locomo non-applicable persons determined by a hospital diagnosis or the like, and an evaluation model is created based on the collected results. It is constructed and used in the motor function evaluation system of this embodiment.

Hereinafter, the motor function evaluation system according to the present embodiment will be described with reference to the drawings. In each of the figures referred to in the following description, the same parts as those of the other figures are indicated by the same reference numerals.

(Structure of motor function evaluation system)
FIG. 1 is a diagram showing an example of a functional block of the motor function evaluation system according to the present embodiment. The motor function evaluation system 1 according to the present embodiment includes a motor function evaluation device 10 and a measuring device 20. In the motor function evaluation system 1 according to the present embodiment, the motor function evaluation device 10 analyzes the measurement data of a person (subject) wearing the measuring device 20 during walking, and estimates whether or not the person is a locomotive risk person. It is a system for.

(Measuring device)
The measuring device 20 measures the change in a specific body part of the subject wearing the device during walking. Examples of the measurement regarding the change in the body part during walking include the measurement of the change in the position of the body part during walking, the change in acceleration, the floor reaction force, and the like. The measuring device 20 includes a sensor unit 210 capable of performing such measurement and outputting measurement data.

In the present embodiment, the measuring device 20 is a wearable measuring device. As a result, the subject can easily wear the measuring device 20, and the locomotive risk can be easily evaluated. Specifically, the measuring device 20 may be, for example, an acceleration sensor, a gyro sensor, a pressure sensor (insole type), or the like. Although only one measuring device 20 is shown in FIG. 1, there are a plurality of measuring devices 20 in the present embodiment.

(Motor function evaluation device)
The motor function evaluation device 10 includes a data analysis unit 110, a storage unit 120, an operation unit 130, and a display unit 140.

The data analysis unit 110 receives a plurality of measurement data acquired by the plurality of sensor units 210 of the measurement device 20 via a network or the like, and executes a process for evaluating the locomotive risk of the subject. The data analysis unit 110 includes a data acquisition unit 112, a locomotive degree score estimation unit 114, and a result output unit 116.

The data acquisition unit 112 acquires a plurality of measurement data acquired by the plurality of sensor units 210 of the measurement device 20 via a network (wired / wireless) or the like. Further, the measurement data may be acquired via, for example, a removable recording medium or another device. The plurality of measurement data acquired by the data acquisition unit 112 is stored in the measurement data storage unit 122 of the storage unit 120.

The locomotive degree score estimation unit 114 has a plurality of measurement data acquired by the data acquisition unit 112 and a plurality of estimation formula data for locomotive risk evaluation stored in the estimation formula storage unit 124 of the storage unit 120. It has a function of estimating (calculating) the locomotive risk of a subject by calculating the variation of the measurement data of the above or the data derived from a plurality of measurement data. The variability of the measured data or the data derived from the measured data is, for example, the variance, standard deviation, quadrant distance, deviation value, range, mean difference, mean of the measured data or the data derived from the measured data. Absolute deviation, discrete entropy. Only one of these may be used or a combination of several may be used in assessing a subject's locomotive risk. Further, the variation of the measurement data or the data derived from the measurement data may be not only the variation of the measurement data itself but also the variation of the data derived (calculated) from the measurement data. For example, acceleration data may be calculated from the movement data of the subject during walking measured by a motion capture system, and the locomotive risk of the subject may be estimated using the variation of the acceleration data.

The result output unit 116 outputs the result data indicating the locomotive risk of the subject estimated by the locomotive degree score estimation unit 114 to the result display unit 144 of the display unit 140. Further, instead of outputting the result data to the result display unit 144, the result data may be output to another device via a network or the like, or the result data may be output to a detachable recording medium or the like. good.

The storage unit 120 has a function of storing various types of data. The storage unit 120 includes a measurement data storage unit 122 and an estimation type storage unit 124.

The measurement data storage unit 122 stores a plurality of measurement data output from the plurality of sensor units 210 of the measurement device 20 via the data acquisition unit 112 of the data analysis unit 110. Of course, the measurement data output from the sensor unit 210 of the measuring device 20 may be stored after being subjected to specific processing.

The estimation formula storage unit 124 stores the estimation formula required for calculating the locomotive risk of the subject. This estimation formula is prepared prior to the subject's locomotive risk assessment (details will be described later).

The operation unit 130 has a function of accepting a user's operation. Specifically, it is an input device such as a keyboard or a mouse. The motor function evaluation device 10 is made to perform various operations such as starting measurement, recording data, and displaying recorded data according to the contents of user input received by the operation unit 130. It may be. Further, the operation unit 130 may output and display the input contents from the user received via the input device or the like on the input information display unit 142 of the display unit 140.

The display unit 140 has a function of displaying and outputting various types of information to the user on the display. The display unit 140 includes an input information display unit 142 and a result display unit 144.

The input information display unit 142 displays the content of the user input received by the operation information input unit 132 of the operation unit 130. The operation unit 130 and the display unit 140 may be integrally configured like a touch panel.

The result display unit 144 acquires the result data indicating the locomotive risk of the subject from the result output unit 116 of the data analysis unit 110 and outputs the result data to a display or the like. Although the output of the result data is to be displayed and output in the present embodiment, it is merely an example and is not limited to this. For example, it may be output by voice or light.

In FIG. 1, the motor function evaluation device 10 and the measurement device 20 are shown as separate devices. For example, the measuring device 20 is a sensor device or the like that can be attached to a plurality of specific parts on the body of the subject, and the motor function evaluation device 10 is a smartphone, a personal computer, or the like, and is measured by the plurality of measuring devices 20. Each measurement data may be transmitted to the motor function evaluation device 10 by wireless communication or the like. Further, the motor function evaluation device 10 and the measurement device 20 may be integrated into one device. For example, it is possible to use a (wearable) portable terminal having a built-in acceleration sensor or the like, such as a smartphone. Furthermore, it can be implemented as a smartphone application (computer program) with a built-in acceleration sensor.

The functional configuration shown in FIG. 1 is merely an example, and is not limited thereto.

(Hardware configuration)
The motor function evaluation system 1 (or motor function evaluation device 10) according to the present embodiment of FIG. 1 can be realized by a hardware configuration similar to that of a general computer device.

FIG. 2 is a diagram showing an example of the hardware configuration of the motor function evaluation system (or motor function evaluation device 10) according to the present embodiment. The computer device 30 shown in FIG. 2 is, for example, a processor 302, a RAM 304, a ROM 306, a hard disk device 308, a removable memory 310, a communication interface (wired / wireless) 312, a display / touch panel 314, and a speaker. It includes a 316 and a keyboard / mouse 318. The functions of each configuration of the motor function evaluation system 1 (or motor function evaluation device 10) described above with reference to FIG. 1 can be realized by the cooperation of hardware and software. For example, the processor 302 can realize the functions of each configuration by reading the program stored in the hard disk device 308 in advance into the memory and executing the program. The hardware configuration shown in FIG. 2 is merely an example, and is not limited thereto.

(Flowchart: Preparation stage)
FIG. 3 is a diagram showing an example of a flowchart of a preparatory stage for performing a locomotive risk evaluation according to the present embodiment. In such a preparatory stage, a plurality of measurement data during walking are collected as sample data for each of the persons who are determined to be locomotive-applicable persons or non-applicable persons by a known method (for example, diagnosis in a hospital). Then, based on the collected sample data, an estimation formula for estimating the locomotive risk is created. In the following description, acceleration data will be adopted as an example of measurement data.

First, in order to collect sample data, a sample locomotive applicable person and a non-applicable person (hereinafter referred to as a subject) each start walking. At this time, a measuring device (accelerometer) is attached to a specific body part of the subject who wants to acquire measurement data as sample data. When the subject starts walking, the measuring device starts recording the measurement data (acceleration data) of the wearing site (step S102). As the measuring device, the measuring device 20 shown in FIG. 1 may be used.

While the subject is walking, acceleration data at the wearing site of the subject to which the measuring device is attached is acquired in chronological order at predetermined time intervals (step S104). The acquired acceleration data is subjected to a low-pass filter and a time rate standardization process of 100% is performed in one walking cycle. The sample data thus obtained is stored in the measurement data storage unit 122 of the storage unit 120 of the motor function evaluation device 10.

Here, the measurement site to be measured may be determined as a measurement site of some of the sites on the body of the subject to which the wearable measuring device can be attached. For example, data showing changes in acceleration at the wearing site in a plurality of steps is acquired. However, it suffices if data showing the change in acceleration in at least one walking cycle can be acquired. Further, the acceleration data to be acquired may be, for example, each component (x, y, z) in the left-right direction, front-rear direction, and vertical direction, or may be configured to acquire only the vertical direction component.

Further, in this example, the acceleration sensor is attached to the subject and the acceleration data is directly acquired from the sensor, but the acceleration data may be acquired by another method. For example, acceleration data may be acquired using a motion capture system. Specifically, for example, a plurality of markers are attached to the body of the subject and walked, and the coordinates of each marker during walking are measured. Then, the measured coordinates may be differentiated twice to calculate the acceleration.

A discrimination model (estimation formula) is generated by a machine learning method based on the acceleration data acquired as described above and the teacher signals of known Locomo applicable persons and non-applicable persons (step S106). The generated discrimination model data is stored in the estimation formula storage unit 124 in the storage unit 120 of the motor function evaluation device 10 (step S108).

(Generation of discrimination model)
Here, an example of generating a discrimination model (estimation formula) is shown. The inventor of the present invention attempted to construct a discrimination model as follows.

The subjects were 75 healthy adults aged 50 years or older (29 locomotive-affected persons, 46 non-locomotive-affiliated persons, average age and distribution 66.0 ± 5.3 years old). These subjects are those who have been previously determined by existing techniques (diagnosis at a hospital, etc.) whether they are Locomo applicable persons or non-applicable persons. The measuring device was attached to the pelvis and trunk of each subject. When the subject starts walking, acceleration data during walking in the pelvis and trunk are acquired in chronological order, the acquired acceleration data is subjected to a low-pass filter, and a time rate standardization process of 100% is performed in one walking cycle. rice field. Then, a representative value indicating the variation for 5 trials was used as a feature quantity (variance, standard deviation, quartile distance, deviation value, range, mean difference, mean absolute deviation, and discrete entropy were obtained).

FIG. 4 is an example of the evaluation result. The "actual measurement value" is a value of walking data that serves as a teacher signal, and is the number of locomotive applicable persons (High Risk) and non-applicable persons (Low Risk) determined in advance for each subject (locomotive applicable person =). 23 + 6 = 29, Locomo non-applicable person 18 + 28 = 46).

The "predicted value" is an estimated value estimated at the time of model construction. The number of people whose measured value was "High Risk" and the predicted value was "High Risk" (23 people), and the number of people whose measured value was "Low Risk" and whose predicted value was "Low Risk" (28 people). Is the number of people who could be evaluated correctly by the estimation formula.

On the other hand, the number of people (6 people) whose measured value was "High Risk" but the predicted value was "Low Risk", and the predicted value was "High Risk" even though the measured value was "Low Risk". The number of people (18 people) is the number of people who could not be evaluated correctly by the estimation formula.

In addition, "sensitivity" is the percentage of people who are actually locomotive-applicable persons who can be correctly evaluated as locomotive-applicable persons (23 / (23 + 6) ≒ 79.3%), and "specificity" is actually locomotive-non-applicable persons. It is the ratio (28 / (18 + 28) ≈ 60.9%) that can correctly evaluate the person who is a person who is not applicable to Locomo.

The "positive predictive value" is the percentage of those who were predicted to be locomotive-affected and who were actually locomotive-affected (23 / (23 + 18) ≈ 56.1%), and the "negative predictive value". Shows the percentage of those who were predicted to be non-locomotives and who were actually non-locomotives (28 / (6 + 28) ≈ 82.4%). Such result data is acquired for the pelvis and trunk. Based on the above trial results, the weighting coefficient and threshold value of the estimation formula are adjusted to determine an appropriate discrimination model. An example of the discrimination model is shown in FIG. In the formula of FIG. 5, "AccXX" represents the XXth acceleration data.

(Flowchart: Preparation stage)
FIG. 6 is a diagram showing an example of a flowchart for performing locomotive risk evaluation in the motor function evaluation system according to the present embodiment. The subject attaches the measuring device 10 to each of a plurality of specific parts (pelvis part and trunk part in the example of FIG. 4) on his / her body and starts walking. When the subject starts walking, the motor function evaluation system 1 starts recording the measurement data (acceleration data) of each wearing site (step S202).

While the subject is walking, the measuring device 10 measures the acceleration of the wearing site at predetermined time intervals (step S204). In addition, the acquired acceleration data is subjected to the same processing as in the preparation stage. That is, in the above example, the acquired acceleration data is subjected to a low-pass filter, and a time rate standardization process of 100% is performed in one walking cycle. The processed acceleration data is sequentially stored in the measurement data storage unit 122.

The locomotive degree score estimation unit 114 reads out the estimation formula stored in the estimation formula storage unit 124 (step S206). The locomotive degree score estimation unit 114 inputs the acceleration data measured and processed in step S204 into the estimation formula and calculates it (step S208). Based on the calculation result, it is determined whether the subject is a Locomo applicable person or a non-corresponding person (step S210). Specifically, for example, the measured acceleration data is inserted into the estimation formula as shown in FIG. 5, and if the calculation result is, for example, "0" or more, the person is not applicable to Locomo. Judge that it is the applicable person. Further, in the present embodiment, acceleration data is acquired from a plurality of measuring devices 10 mounted on a plurality of parts on the body of the subject. Then, based on these acceleration data, it is determined whether or not the locomotive is applicable. For example, it is calculated by an estimation formula for each acceleration data of each part, and if the ratio of parts whose calculation result is "0" or more exceeds a predetermined threshold, it is judged to be Locomo applicable. If the calculation result of the part is "0" or more, it may be determined that the person is not a Locomo non-applicable person, and so on.

The determination result in step S210 is output from the result output unit 116 and displayed on the result display unit 144 (step S212). Of course, the judgment result may be output not only by display but also by voice or light, for example.

Although one embodiment of the present invention has been described so far, it goes without saying that the present invention is not limited to the above-described embodiment and may be implemented in various different forms within the scope of the technical idea.

Also, the scope of the present invention is not limited to the exemplary embodiments illustrated and described, but also includes all embodiments that provide an effect equal to that intended by the present invention. Furthermore, the scope of the present invention is not limited to the combination of the features of the invention defined by each claim, but may be defined by any desired combination of the specific features of all the disclosed features. ..

1 Motor function evaluation system 10 Motor function evaluation device 20 Measuring device 110 Data analysis unit 120 Storage unit 130 Operation unit 140 Display unit 210 Sensor unit

Claims (4)

It is a motor function evaluation system that evaluates the risk of locomotive syndrome and feeds it back to the user as the motor function of the subject.
A plurality of wearable sensor devices worn on a plurality of specific parts of the subject's body, and a plurality of wearable sensor devices.
A data acquisition unit that acquires a plurality of measurement data indicating changes in a plurality of specific parts of the body of the subject in a walking motion from the plurality of wearable sensor devices.
To calculate the variation of the plurality of measurement data or the variation of the data derived from the plurality of measurement data, and to estimate the locomotive risk of the subject by using the calculated value of the variation corresponding to each of the plurality of specific sites. The ratio of the parts whose output of the estimation formula is equal to or higher than the first threshold value is obtained from the plurality of specific parts, and the ratio of the parts having the output of the first threshold value or more is the second. When the threshold value is exceeded, the locomotive degree score estimation unit that determines that the subject is locomotive-corresponding, and the locomotive degree score estimation unit.
As the estimation result of the locomotive risk of the subject, a result output unit that outputs whether or not the locomotive is applicable,
Have,
Motor function evaluation system. The variation of the plurality of measurement data is any one of the variance, standard deviation, quartile distance, deviation value, range, mean difference, mean absolute deviation, discrete entropy, or a combination thereof of the plurality of measurement data. ,
The variation of the data derived from the plurality of measurement data includes the dispersion, standard deviation, quartile distance, deviation value, range, mean difference, mean absolute deviation, and discrete entropy of the data derived from the plurality of measurement data. Any of, or a combination of these,
The motor function evaluation system according to claim 1. A motor function evaluation method executed by a computer system having a plurality of wearable sensor devices mounted on a plurality of specific parts of the subject's body, a data acquisition unit, a locomotive degree score estimation unit, and a result output unit. And
A step in which the data acquisition unit acquires a plurality of measurement data indicating changes in a plurality of specific parts of the body of the subject in a walking motion from the plurality of wearable sensor devices.
The locomotive degree score estimation unit calculates the variation of the plurality of measurement data or the variation of the data derived from the plurality of measurement data, and calculates the value of the variation corresponding to each of the plurality of specific parts. It is applied to the estimation formula for estimating the locomotive risk of the subject, and the ratio of the parts where the output of the estimation formula is equal to or higher than the first threshold value is obtained from the plurality of specific parts, and the value is equal to or higher than the first threshold value. A step of determining the subject as a locomotive when the proportion of the site becomes equal to or higher than the second threshold value, and
A step in which the result output unit outputs whether or not the locomotive is applicable as the estimation result of the locomotive risk of the subject.
Motor function evaluation methods, including. A computer program for causing the computer system to execute the motor function evaluation method according to claim 3.

Images