JP2019084130A - Walking motion evaluation apparatus, walking motion evaluation method, and program - Google Patents

Abstract

An object of the present invention is to accurately evaluate walking motion without being influenced by changes in walking speed and the like.
At least an angle and an angular velocity of a joint of a lower leg at the time of walking of a person to be evaluated are acquired as a walking motion variable. Then, the acquired walking motion variable is adjusted based on the walking speed of the person to be evaluated, and regression analysis is performed using the adjusted walking motion variable as an explanatory variable and age as a target variable. From the result of regression analysis by this regression analysis process, an index of the degree of aging of the walking motion of the subject is obtained.
[Selected figure] Figure 9

Description

  The present invention relates to a walking motion evaluation device and a walking motion evaluation method, and a program for performing evaluation of walking motion.

  Humans acquire single walking (bipedal walking) at the age of just over one year, and then, after adulthood and after middle age, the walking behavior changes due to the influence of aging. This is a comprehensive result of age-related changes in physical function. It is important to know the change of the walking movement with age after adulthood, to pursue its kinematic cause, and to search for measures for maintenance and improvement of the walking movement and walking ability based on these evidences. It is an extremely important issue to satisfy “long life”.

  Conventionally, for example, a technique described in Patent Document 1 is known as an evaluation method regarding walking. Patent Document 1 describes a method of measuring a walking action using a sheet-type pressure sensor and measuring a walking speed with an acceleration sensor to evaluate a walking age. In the method described in this patent document 1, using the right walking angle, the left walking angle, the walking cycle, the both leg supporting period time, and the stride as the walking parameters, the walking age is obtained by the multiple regression equation including these as explanatory variables. Is calculated.

JP 2014-94070 A

  As described in Patent Document 1, various evaluation methods for estimating the age by measuring the walking action have been conventionally proposed, but a more advanced evaluation of the age and the degree of aging has been required. That is, conventionally proposed evaluation methods do not take into consideration that the walking motion changes depending on the walking speed. Therefore, it is difficult to say that the evaluation result obtained is greatly different from the individual-specific accurate evaluation if the pedestrian walks at a slightly faster or slower speed, and the evaluation result greatly changes.

  An object of the present invention is to provide a walking motion evaluation device, a walking motion evaluation method, and a program capable of accurately evaluating a walking motion without being affected by a change in walking speed or the like.

  A walking motion evaluation device according to the present invention includes a walking motion variable acquisition unit that acquires at least an angle and an angular velocity of a joint of a lower limb during walking of a person to be evaluated as a walking motion variable, and a walking motion acquired by the walking motion variable acquisition unit An adjustment unit that adjusts a variable based on the walking speed of a person to be evaluated, a regression analysis unit that performs regression analysis using a walking motion variable adjusted by the adjustment unit as an explanatory variable and age as an objective variable, and a regression analysis unit And an age index acquisition unit for acquiring an index of the aging degree of the walking motion of the person to be evaluated from the result of the regression analysis.

  Further, according to the walking motion evaluation method of the present invention, the walking motion variable acquisition processing for acquiring at least the angle and the angular velocity of the joints of the lower limbs at the time of walking of the evaluation target person as walking motion variables Adjustment processing for adjusting a walking motion variable based on the walking speed of a person to be evaluated, regression analysis processing for performing regression analysis using a walking motion variable adjusted by the adjustment processing as an explanatory variable and age as a target variable It includes an age index acquisition process for acquiring an index of the aging degree of the walking motion of the person to be evaluated from the result of the regression analysis by the analysis process.

  Further, the program of the present invention causes a computer to execute a procedure for executing each of the walking motion variable acquisition processing, adjustment processing, regression analysis processing, and degree-of-age index acquisition processing of the above-described walking motion evaluation method.

  According to the present invention, it is possible to indicate the degree of aging of the walking motion, and to support the maintenance and improvement of the walking ability and motion of the individual. This will help elderly people to keep walking well no matter how old they are, and will contribute to the reduction of medical expenses and the realization of a healthy longevity society. In addition, elderly people, as well as middle-aged and elderly people who will become elderly in the future, can also be educated about maintenance and improvement of walking motion. Furthermore, the evaluation of the degree of age according to the present invention can also contribute to the development of a training method for improving walking motion.

It is a functional block diagram showing an example of composition of a walk movement evaluation system by a 1 embodiment example of the present invention. It is a flow chart which shows an example of walking movement evaluation processing by one example of an embodiment of the present invention. It is a figure which shows the example of the torque and angle of the hip joint according to one embodiment of this invention, a knee joint, and an ankle joint. It is a figure which shows the example of the angle of a hip joint at the time of walking of 1 cycle (two steps), angular velocity, torque, and torque power. It is a figure which shows the example before adjustment (FIG. 5A) of a walking speed and a height, and after adjustment (FIG. 5B). It is a figure which shows the outline | summary of the walking motion age-degree index by one example of embodiment of this invention. It is a figure which shows the example of the age-related index of walk operation | movement using Z score by one embodiment of this invention. It is a figure which shows the example of distribution of the age-related index of the walking motion by one example of embodiment of this invention. It is a figure which shows an example of the profile of the age-related-index of walk operation | movement by one embodiment of this invention. It is a figure which shows an example of the multiple regression analysis result (when it evaluates from kinematics variable and kinetic variable) of age prediction by one embodiment of this invention. It is a figure which shows the peak value definition of the joint angle applied to the example of one embodiment of this invention. It is a figure which shows the peak value definition of the joint angular velocity applied to the example of one embodiment of this invention. It is a figure which shows the peak value definition of the joint torque applied to the example of one embodiment of this invention. It is a figure which shows the peak value definition of joint torque power applied to the example of one embodiment of this invention. It is a figure which shows another example of the profile of the age-related-index of walk operation | movement by one example embodiment of this invention. It is a figure which shows an example of the multiple regression analysis result (when it evaluates only from a kinematics variable) of the age prediction by one embodiment of this invention.

  Hereinafter, an embodiment of the present invention (hereinafter, referred to as "this example") will be described with reference to the attached drawings.

[1. Configuration of walking motion evaluation device]
FIG. 1 is a functional block diagram showing the configuration of the walking motion evaluation device 10 of the present example. The walking motion evaluation device 10 of this example is configured, for example, by mounting a program for executing walking motion evaluation processing on a computer device.

  The walking motion evaluation apparatus 10 shown in FIG. 1 acquires the angle and torque of each joint of the lower leg (leg) at the time of walking of the person to be evaluated. Specifically, the walking motion evaluation apparatus 10 includes a hip joint information acquiring unit 11, a knee joint information acquiring unit 12, and an ankle joint information acquiring unit 13, and the respective acquiring units 11 to 13 determine the angles of the joints and Get the torque. Further, each acquisition unit 11 to 13 acquires an angular velocity obtained by differentiating the angle of the joint with respect to time. Furthermore, each acquisition unit 11 to 13 acquires torque power obtained by multiplying the torque of the joint by the angular velocity. Although details will be described later, only part of the information (angle, angular velocity, torque, and torque power) may be acquired in evaluating the walking motion.

  Each acquisition part 11-13 acquires the angle of a joint from the image | video which image | photographed the walking state of the to-be-evaluated person. Moreover, each acquisition part 11-13 acquires a torque from the ground reaction force which the power meter laid on the floor of the location which a to-be-evaluated person walks detects. As a force meter here, it is preferable to use a sensor called a force platform which can obtain triaxial force applied to the foot from the ground. However, the use of a force platform is an example, and the torque of each joint may be obtained from another force meter (for example, a pressure sensor that detects a pressure distribution). Furthermore, the angle or the torque of the joint may be obtained from a sensor attached to the lower leg of the evaluation subject. Details of the torques and angles of these joints will be described later.

The information on each joint acquired by each acquisition unit 11 to 13 is supplied to the walking motion variable acquisition unit 21. The walking motion variable acquisition unit 21 supplies the information (angle, angular velocity, torque, and torque power of each joint) to the walking speed / height adjusting unit 22 as a walking motion variable. The walking speed and height adjustment unit 22 determines the supplied walking motion variable based on the walking speed of the evaluation target person detected by the walking speed detection unit 14 and the height of the evaluation target person detected by the height detection unit 15. adjust. Details of the adjustment process in the walking speed / height adjusting unit 22 will be described later.
The detection of the walking speed by the walking speed detection unit 14 and the detection of the height of the evaluation subject by the height detection unit 15 are performed based on, for example, an image obtained by photographing the walking state of the evaluation subject. Alternatively, the walking speed and the height may be directly input using a keyboard or the like.

  Information on the walking motion variables (angle, angular velocity, torque, and torque power of each joint) whose walking speed and height have been adjusted by the walking speed / height adjusting unit 22 is supplied to the regression analysis unit 23. The regression analysis unit 23 performs regression analysis with the supplied walking motion variable as an explanatory variable and age as a target variable. The regression analysis unit 23 of this example performs stepwise multiple regression analysis known as one method of regression analysis.

  Then, the regression analysis unit 23 supplies the information of the age that is the objective variable obtained by the regression analysis to the walking motion age index acquisition unit 24. The walking motion age index acquisition unit 24 obtains walking motion age index using information on age. When the walking motion age index acquisition unit 24 of this example obtains the walking motion age index, the walking motion age index is obtained using the Z score which is one of the analysis methods.

  Information on the walking motion age index acquired by the walking motion age index acquisition unit 24 is supplied to the display data creation unit 25. The display data creation unit 25 creates display data for displaying the supplied walking motion age index, and displays the walking motion age index using the created display data.

[2. Flow of the entire walking motion evaluation process]
FIG. 2 is a flowchart showing an outline of a flow of processing when the walking motion evaluation device 10 executes the walking motion evaluation processing.
First, the walking motion variable acquisition unit 21 performs walking motion variable acquisition processing for acquiring the angle, the angular velocity, the torque, and the torque power for each of the hip joint, the knee joint, and the ankle joint (step S1).
Then, the angle, angular velocity, torque, and torque power of each joint acquired by the walking motion variable acquisition unit 21 are adjusted by the walking speed / height adjustment unit 22 based on the walking speed and height of the evaluation subject ( Step S2).

Thereafter, the regression analysis unit 23 performs regression analysis by stepwise multiple regression analysis (step S3). At this time, the regression analysis unit 23 performs regression analysis processing with the walking motion variables (angle, angular velocity, torque, and torque power) whose walking speed and height have been adjusted as explanatory variables and age as a target variable.
Furthermore, based on the result of regression analysis by the regression analysis unit 23, the walking motion age index acquisition unit 24 obtains a walking motion age index using the Z score (step S4). Then, the display data creation unit 25 performs processing for creating display data based on the walking motion age index obtained by the walking motion age index acquisition unit 24 and further evaluation using the walking motion age index Output processing of the result is performed (step S5).

[3. Example of torque and angle to be acquired]
FIG. 3 shows an example of the torque and angle of each joint acquired by the acquisition units 11 to 13.
In the case of this example, as shown in FIG. 3, in order to analyze the walking behavior of the evaluation target person, a hip torque T _H, and hip joint angle theta _H, and knee joint torque T _K, the knee joint angle theta and _K, and the ankle joint torque T _A, and the ankle joint angle theta _A is obtained.
Hip joint angle theta _H is an angle formed by the trunk and thigh, hip torque T _H is the torque produced by the hip.

The knee joint angle θ _K is an angle formed by the thigh and the lower leg, and the knee joint torque T _K is a torque generated at the knee joint.
The ankle joint angle theta _A, the angle formed by the lower leg and foot (before ankle), ankle torque T _A is the torque generated by the ankle joint.

The angles θ _H , θ _K , θ _A of the joints can be differentiated with respect to time to obtain angular velocities ω _H , ω _K , ω _A. Further, torque powers P _H , P _K , and P _A can be obtained by multiplying the torques T _H , T _K , and T _{A of the} joints and the angular velocity.

FIG. 4 shows an example of the acquisition state of angle and torque at the time of walking.
Each graph shown in FIG. 4 is for a hip joint at the time of walking (the hip joint will be described as an example, but similar graphs and feature points are also acquired at the knee joint and the ankle joint). Here, as shown in FIG. 4A, the movement in one cycle (two steps) of one foot (right foot or left foot) is shown normalized to 100%. The horizontal axes of FIGS. 4B to 4E are normalized values from 0% (timing to ground) to 100% (timing to next grounding).

FIG. 4B shows an example of change of the hip joint angle θ _H in one cycle. The vertical axis in FIG. 4B indicates the value of the angle.
During walking, the hip joint angle θ _H is the largest angle near the position where the foot leaves the ground, and the smallest angle near the position where the distant foot touches the ground. Here, the walking motion age index acquisition unit 24 of the present example acquires a peak value among changes in the hip joint angle θ _{H in} each cycle. Specifically, within one cycle period shown in FIG. 4B, the largest angle θ _H1 and the smallest angle θ _H2 are acquired as feature point information.

FIG. 4C shows an example of change in the hip joint angular velocity ω _H in one cycle. The vertical axis | shaft of FIG. 4C shows the value of angular velocity.
The hip joint angular velocity ω _H has a period in which the value in the positive direction and a period in which the value is in the negative direction during one cycle. Here, the walking motion age index acquisition unit 24 according to the present example uses three angular velocities ω _H1 , ω _H2 , and ω _H3 as peak or bottom in the hip joint angular velocity ω _{H in} each cycle as feature point information. get.

FIG. 4D shows an example of change of the hip joint torque T _H in one cycle. The vertical axis in FIG. 4D indicates the value of torque.
For even hip torque T _H, during a cycle, there is a period in which a positive-direction value to become period and negative values. Here, the walking motion age index acquisition unit 24 of this example uses three torques T _H1 , T _H2 , T _H3 that become peaks or bottoms as the feature point information in the hip joint torque T _{H in} each cycle. get.

Figure 4E shows a variation of a single cycle of the hip torque power P _H. The vertical axis | shaft of FIG. 4E shows the value of torque power. For even hip torque power P _H, during a cycle, there is a period in which a positive-direction value to become period and negative values. Here, the walking motion age index acquisition unit 24 of the present example determines four torque powers P _H1 , P _H2 , P _H3 , and P _H4 that become peaks or bottoms in the hip joint torque power P _H in each cycle. Is acquired as feature point information.

[4. Adjustment of walking speed and height]
FIG. 5 shows an example of the adjustment state performed by the walking speed / height adjustment unit 22. FIG. 5A shows the state before adjustment and FIG. 5B shows the state after adjustment. In the example of FIG. 5, the vertical axis represents the predicted value of the fifth feature point T _K5 of the knee joint torque T _K , and the horizontal axis represents the observed value of the knee joint torque. The meaning of the fifth feature point T _K5 of the knee joint torque T _K will be described later.
The adjustment is performed by the following [Equation 1]. The definition of each value shown in the formula 1 is also shown below.

The observed values on the horizontal axis in FIGS. 5A and 5B are observed values including the effects of walking speed and height, and in the graph shown in FIG. 5A, change characteristics of observed values and predicted values a due to the effects of walking speed and height. Increases linearly according to walking speed and height. On the other hand, in the graph shown in FIG. 5B where the adjustment shown in [Equation 1] is performed, the change characteristic b of the observed value and the predicted value is constant, and the observed value in which the influence of walking speed and height is excluded It has become.
In addition, although adjustment was performed by two, walking speed and height here, you may make it the walking speed and height adjustment part 22 adjust only walking speed, for example.

[5. Evaluation of Aging Change of Walking Action]
Next, an aging change of the walking motion will be described.
Specific characteristics of age-related change in walking motion are "low cadence", "short swing phase", "small head up and down movement", "small pelvic rotation", "late support period late Small plantar flexion, "small knee flexion and jerk on the first half of the swing phase,""low toe on the end of the swing phase,""small foot angle at touch,""shoulder"Bending is small", "extending of elbow is small", "a large variation in head side-to-side movement", "a large minimum value of clearance in the swing phase (distance between toe and ground)" and the like can be mentioned.

  Since these changes in walking motion also depend on the walking speed, the characteristics of the walking motion of the elderly can be thought to be a combination of changes in walking motion due to aging and changes in motion due to slow walking . When walking at the same speed as a young person, the walking action of the elderly is “short step length and high step frequency”, “more knee and hip joints in flexion through walking cycle, more ankle joints dorsiflexion The characteristics are as follows: "the anteroposterior tilt of the trunk is large", "the maximum plantarflexion is small", and "the joint angular velocity of the knee joint and hip joint during the swing phase is large".

When age-related changes in lower extremity joint movement during walking are viewed from the joint torque power, in the elderly, the positive ankle joint torque power in the second half of the support period is reduced to about two thirds of the young. The joint torque power here is a value obtained by multiplying the joint torque by the joint angular velocity.
The positive ankle joint torque power in the second half of the support period is mainly due to the foreshortening contraction of the plantar flexor muscles and plays a role in accelerating the body forward and increasing the step length. For this reason, it is considered that the decrease in ankle torque power in the second half of the supporting period with aging is a cause of shortening of the step length of the elderly.

  On the other hand, the negative power of the knee joint during the swing phase and the positive power of the hip tend to be greater in the elderly than in the young. During walking, the knee joint flexes and separates and continues to flex to the middle of the swing phase, then it extends in the late swing phase and slightly flexes again to reach the ground. The hip joint flexes and leaves, and continues to flex until the mid swing phase, and it extends and reaches the ground in the late swing phase. The large joint torque power associated with these lower-limb movements makes it possible to accelerate the flexion and extension movements of the knee joint and the hip joint and to increase the step frequency. When walking at the same speed, the elderly walk with a shorter step length and a higher step frequency than young people because of the characteristics of these lower extremity joint power exertion.

Based on the age-related change of the walking motion as described above, the walking motion evaluation device 10 of this example is configured to evaluate the aging degree of the walking motion.
Conventionally, when evaluating the change in walking motion accompanying aging, since the individual difference in walking motion (inter-individual variation) is large, and because there is no standard value that serves as a reference for evaluation, aging of the walking motion It was difficult to assess the degree appropriately.
On the other hand, in the case of this example, one of the features is that the exercise technique is evaluated in consideration of the variation in movement among individuals.

First, 177 healthy subjects from 19 to 82 years are subjects (here all women), and depending on the age, young people group (32 over 19 years old and under 50 years old), middle aged people group (50 to 65 years old) Divided into 3 groups of less than 36 people, elderly people group (109 people over 65 years old). The young people group FY, the middle-aged people group FM, and the old people group FE here are shown in FIG. 8 described later.
The walking motions of these subjects were measured, and the kinematics (joint angle and joint angular velocity) and kinetics (joint torque per body weight and joint torque power) of the three lower limbs joints (foot, knee and hip joint) were calculated. After that, 43 feature points (peak values) of each parameter were extracted in total. The process of extracting feature points is performed by the walking motion variable acquisition unit 21.

Next, with reference to FIG. 6, the walking motion age index will be described.
First, the walking speed / height adjustment unit 22 adjusts the above-mentioned 43 feature point data with the walking speed and the height. Then, the regression analysis unit 23 performs stepwise multiple regression analysis using the feature point data adjusted by the walking speed / height adjustment unit 22 as an explanatory variable and age as a target variable (step S11).

Among the explanatory variables, those which do not meet the following two conditions were excluded from the multiple regression analysis (step S12).
1. The correlation with age is significant.
2. Less than 50% variation (rms _CV iFY) in the reference group (here the young people group).
The predicted age obtained by this multiple regression analysis is converted into a Z score, which is defined as a walking motion age index (GMA index; Gait-Motion Aging index) (step S13).

Note that the Z score for converting the predicted age is obtained by scoring the degree of deviation from the individual's standard group average in consideration of the variation of the standard group.
That is, as shown in FIG. 7, the average of the reference group is compared with the value of the subject (subject) to calculate the deviance degree, and the walking motion evaluation is performed using the z-score that quantifies the deviance degree. To do. Here, the Z score is calculated from the following [Equation 2]. In the equation (2), y is the value of the subject (subject), y _c (̄) is the mean value of the reference group, and SD _c is the standard deviation of the reference group. Note that (̄) is a symbol originally written on the immediately preceding code y _c as shown in Equation 2, but is denoted as y _c (̄) due to a notational restriction.

  Here, since the reference group is the young people group (FY) in order to evaluate the aging degree of the walking movement, the obtained walking movement aging degree index (GMAindex) comprehensively deviates from the young people group It becomes an index to evaluate.

FIG. 8 shows the relationship between the obtained walking motion age index (GMA index: vertical axis) and age (age: historical axis: horizontal axis).
Here, the closer to 0 the walking motion age index (GMAindex) indicates that the walking motion closer to the young person is shown, and the larger the walking motion, it means that it deviates from the walking motion of the young person. That is, every increase of 1 means that the standard deviation of the young people group deviates from the average of the young people group.

  As shown in FIG. 8, it can be seen that the walking movement age index (GMAindex) increases with age, but a large deviation is seen between individuals even at the same history age. The regression line x in FIG. 8 indicates a standard value in each age. That is, an individual located below the regression line x walks closer to a younger person than the standard movement of the same age, and an individual located above the regression line x is older than the standard movement of the same age It can be evaluated as walking.

The subject A and the subject B shown in FIG. 8 are both in their late 60's.
In the case of the subject A, the walking motion age index (GMAindex) is close to 0, and it is considered that the walking motion is close to a young person. On the other hand, in the case of the subject B, it is largely deviated from the standard movements of the young person and the 70-year-old, and it can be said that the aging degree of the walking movement is high.

Since the walking motion age index (GMAindex) is obtained by multiple regression analysis, each term of the multiple regression equation can be regarded as a breakdown of the walking motion age index (GMAindex). For this reason, it is possible to identify separately the factor which affects the aging degree of walking motion by examining this breakdown. As a result of stepwise multiple regression analysis, the following five variables remained as each term of the multiple regression equation. The meaning of each variable will be described later (FIGS. 11 to 14).
1. Maximum knee flexion torque just before contact: T _K5 (The flexion torque is greater for the elderly)
2. Ankle plantar flexion angle near the point of departure: θ _A3 (The more the elderly, the smaller the plantar flexion)
3. Maximum hip flexion angle during swing phase: θ _H2 (The flexion is greater for the elderly)
4. Ankle joint maximum plantar flexion torque in the second half of support period: T _A2
5. Maximum value of negative power due to hip flexion torque in late support period: P _H2 (negative power increases as elderly people)

FIG. 9 is a radar chart obtained by converting the five variables T _K5 , θ _A3 , θ _H2 , T _A2 , and P _H2 into Z scores. The z-score for each variable is zero for the young group average (FY), and there is a plot inside the chart the further from the young group. FE is the average of the elderly.
The characteristic Da of the subject A of the elderly is a plot similar to that of the young group, but it is understood that the deviation of the variables T _K5 , θ _A3 and P _H2 is particularly large in the characteristic Db of the subject B of another elderly. It is known that in the elderly, contracture of the hip flexor group and muscle weakness of the plantar flexor group are observed, but the deviation of the variable θ _A3 and the variable PH ₂ observed in the subject B causes these age-related changes. Is considered to be reflected.
As can be seen from, for example, that the ankle joint maximum plantarflexion angle θ _A3 in the example of FIG. 9 is largely different between the characteristics Da and Db of the two people, the information about the ankle joint contains the most important information regarding the aging degree It is important to use information on at least the ankle joint to assess the degree of aging.

  In the radar chart shown in FIG. 9, for example, display data to be displayed by the display data generation unit 25 is generated, and a display unit (not shown) connected to the display data generation unit 25 corresponds to the corresponding radar chart. Display the walking movement age index. In the example of FIG. 9, there are five peaks (important peaks for evaluation) finally remaining as a result of the multiple regression analysis, and the radar chart shows the five peaks as a pentagonal figure. is there. If the number of peaks finally left as a result of the multiple regression analysis is different from this, the walking movement age index will be displayed on a radar chart other than a pentagon. The radar chart of FIG. 15 described later is an example in the case where three other peaks remain as a result of the multiple regression analysis.

FIG. 10 is a table showing the results of multiple regression analysis of age prediction in FIG. That is, FIG. 10 shows the case where the multiple regression analysis shown in FIG. 9 is evaluated from the kinematics variable and the kinetic variable. Here, the correlation coefficient R ^{2 is} 0.269 (p <0.001).
At this time, the predicted age [yrs] can be obtained by the following equation using partial regression coefficients of five peak variables and a constant term.
Predicted age [yrs] = − 66.1409 × T _K5 [N · m / kg] −0.4963 × θ _A3 [deg] −0.4962 × θ _H2 [deg] -19.7917 × T _A2 [N · m / kg] −9.1245 × P _H2 [W / kg] + 207.0688

  11 to 14 show specific examples of peak values of respective values acquired by the walking motion variable acquisition unit 21. FIG. The horizontal axis of each graph in FIGS. 11 to 14 is a value obtained by normalizing one cycle of walking (two steps) of the right foot or the left foot to 100%, and the vertical axis indicates angle, angular velocity, torque, or torque power. . In each graph, a portion indicated by a circle is a peak value acquired by the walking motion variable acquisition unit 21.

FIG. 11 shows the definition of peak values of joint angles. The upper stage shows the hip joint angle, the middle stage shows the knee joint angle, and the lower stage shows the ankle joint angle.
The hip joint angle in the upper part of FIG. 11 sequentially acquires a hip joint extension angle peak θ _H1 (immediately before release) and a hip joint flexion angle peak θ _H2 (middle swing phase) within one walking cycle.
The knee joint angle in the middle of FIG. 11 is, within one walking cycle, knee joint flexion angle peak θ _K1 (immediately after grounding), knee joint extension angle peak θ _K2 (support phase), knee joint flexion angle peak θ _K3 Immediately after the ground, and knee joint extension angle peak θ _K4 (immediately before contact) are acquired in order.
The ankle joint angle in the lower part of FIG. 11 is, within one walking cycle, ankle plantar flexion angle peak θ _A1 (immediately after grounding), ankle dorsiflexion angle peak θ _A2 (late support period), and ankle plantar flexion angle peak Acquire θ _A3 (near off ground) in order.

FIG. 12 shows the definition of the peak value of the joint angular velocity. The upper stage shows the hip joint angular velocity, the middle stage shows the knee joint angular velocity, and the lower stage shows the ankle joint angular velocity.
The hip joint angular velocity in the upper part of FIG. 12 is the hip joint extension angular velocity minimum value ω _H1 (immediately after grounding), the hip joint extension angular velocity peak ω _H2 (first half of support period), and the hip joint bending angular velocity peak ω _H3 ), In order.
The knee joint angular velocity shown in the middle of FIG. 12 is, within one walking cycle, knee joint flexion angular velocity peak ω _K1 (immediately after grounding), knee joint extension angular velocity peak ω _K2 (first half of support period), knee joint flexion angular velocity peak ω _K3 Near the ground) and knee joint extension angular velocity peak ω _K4 (late swing phase) are acquired in order.
Lower leg joint angular velocity in FIG. 12, in one walking period, ankle plantar屈角speed peak omega _A1 (immediately after the ground), ankle dorsiflexion speed peak omega _A2 (immediately after the ground), ankle plantar屈角speed peak omega _A3 (Near the separation), the ankle dorsiflexion angular velocity peak ω _A4 (immediately after the separation), and the ankle plantar flexion angular velocity peak ω _A5 (the late swing phase) are sequentially acquired.

FIG. 13 shows the definition of the peak value of joint torque. The upper stage shows the hip joint torque, the middle stage shows the knee joint torque, and the lower stage shows the ankle joint torque.
Upper hip torque of FIG. 13, in one gait cycle, hip extension torque peak T _H1 (immediately after the ground), hip flexion torque peak T _H2 (Hanarechi immediately before), and hip extension torque peak T _H3 (the ground just before) Get in order.
The knee joint torque in the middle of FIG. 13 is, within one walking cycle, knee joint flexion torque peak T _K1 (immediately after grounding), knee joint extension torque peak T _K2 (immediately after grounding), knee joint flexion torque peak T _K3 (support period The second half), knee joint extension torque peak T _K4 (immediately before release), and knee joint flexion torque peak T _K5 (immediately before contact) are acquired in order.
The ankle joint torque torque T _A1 (immediately after touching the ground) and the ankle plantar flexion torque peak T _A2 (immediately before the release) are sequentially acquired in one walking cycle in the lower ankle joint torque in FIG. 13.

FIG. 14 shows the definition of the peak value of joint torque power. The upper stage shows the hip joint torque, the middle stage shows the knee joint torque, and the lower stage shows the ankle joint torque.
The hip torque in the upper part of FIG. 14 is, within one walking cycle, positive power peak P _H1 (immediately after grounding) by hip extension torque, negative power peak P _H2 by hip flexion torque (later in support period), positive power by hip flexion torque The peak P _H3 (near the release point) and the positive power peak P _H4 (immediately before contact) by the hip joint extension torque are acquired in order.
The knee joint torque in the middle part of FIG. 14 is, within one walking cycle, positive power peak P _K1 (immediately after grounding) by knee joint flexion torque, negative power peak P _K2 (immediately grounding) by knee joint extension torque, knee joint extension torque Positive power peak P _K3 (early supporting period), negative power peak P _K4 (right before detachment) by knee joint extension torque, and negative power peak P _K5 (right before grounding) by knee joint bending torque are acquired in order.
The ankle joint torque in the lower part of FIG. 14 is, within one walking cycle, a negative power peak P _A1 (immediately after grounding) by ankle dorsiflexion torque, a negative power peak P _A2 (early supporting period) by ankle plantar flexion torque, The positive power peak P _A3 (immediately before the release) by the ankle plantar flexion torque is acquired in order.

Among these peak values, in the example of FIG. 9, five peaks T _K5 , θ _A3 , θ _H2 , T _A2 , and P _H2 finally remain as a result of the regression analysis, and the five peaks T _K5 and θ _The walking motion age index (GMAindex) is obtained from _A3 , θ _H2 , TA ₂ and PH ₂ . It is an example that the five peaks shown here finally remain as a value for obtaining the walking movement age index, and the remaining peaks remain as values for obtaining the walking movement aging index according to the state of regression analysis. Sometimes.

  As described above, by evaluating the degree of aging of the walking motion, it is possible to enlighten the maintenance and improvement of the walking motion not only for the elderly but also for middle-aged and young people who will become elderly in the future become. Also, by showing specific improvement points in operation, it becomes possible to devise and implement appropriate training for operation improvement.

[6. Modified example]
In the embodiment described above, the walking motion variable acquisition unit 21 acquires the angle, the angular velocity, the torque, and the torque power for each of the hip joint, the knee joint, and the ankle joint, and performs regression analysis using them. The walking motion age index was obtained from the obtained peaks. The use of these torques and angles is an example, and it is possible to obtain the walking movement age index using only a part of information of these torques and angles.

  For example, the walking motion variable acquisition unit 21 acquires the angle and angular velocity of the hip joint, the angle and angular velocity of the knee joint, and the angle and angular velocity of the ankle joint. After adjustment, the regression analysis unit 23 may perform regression analysis, and the walking motion age index acquisition unit 24 may obtain the walking motion age index.

FIG. 15 shows an example in which the angle and angular velocity of each joint are acquired, and the walking motion age index is acquired from three peaks obtained by regression analysis using them.
In this example, as a result of stepwise multiple regression analysis using the angle and angular velocity of each joint, the following three variables remain as each term of the multiple regression equation.
1. Maximum hip flexion angle in the middle of swing phase: θ _H2
2. Maximum ankle flexion angle near the ground: θ _A3
3. Ankle joint maximum base bending angular velocity just after touch: ω _A1

FIG. 15 is a radar chart obtained by converting the three variables θ _H2 , θ _A3 and ω _A1 into Z scores. Also in the case of FIG. 15, as in FIG. 9, the Z score of each variable is zero at the young group average (FY), and there is a plot inside the chart as it deviates from the young group. FE is the average of the elderly.
In FIG. 15, the score Dc of the subject B and the score Dd of the subject A are shown.

FIG. 16 is a table showing the results of multiple regression analysis of age prediction in FIG. That is, FIG. 16 shows the case where the multiple regression analysis shown in FIG. 15 is evaluated from the kinematics variable. Here, the correlation coefficient R ² = 0.154 (p <0.001).
At this time, the predicted age [yrs] is obtained by the following equation using partial regression coefficients and constant terms of three peak variables.
Predicted age [yrs] = − 0.6134 × θ _H2 [deg] −0.5116 × θ_A3 [deg] −3.3450 × ω_A1 [rad / s] +233.9590

As described above, it is possible to obtain the walking motion age index by using the angle of each joint of the lower leg and the angular velocity obtained by differentiating the angle with time. In the case of the example of FIG. 15, since the predicted age is obtained from the angle and angular velocity of the joint which is the kinematics variable, as compared with the case of evaluation from the kinematics variable and the kinematics variable as in the example of FIG. The accuracy of the predicted age may be slightly worse. However, since the joint angle and angular velocity, which are the kinematics variables, can be acquired only from the video captured by the video camera etc. of the subject's walking condition, acquisition of the walking motion variables is easier than in the example of FIG. it can. Also, in the case of the example of FIG. 15, since each of the joints is evaluated from two variables using the angle of the ankle joint which is an important joint in estimating the degree of aging, an appropriate evaluation is made. Have the effect of
Note that the profile of the walking motion aging degree is shown as a polygonal radar chart shown in FIG. 9 and FIG. 15 as an example, and the profile of the walking motion aging degree is displayed in a figure or graph of another display form. You may do so.

In the embodiment described above, the regression analysis unit 23 performs stepwise multiple regression analysis (linear regression), but the regression analysis unit 23 may perform other regression analysis.
The use of the Z score in obtaining the walking movement age index is also an example, and an index related to the walking movement aging degree may be obtained by another analysis method.

  DESCRIPTION OF SYMBOLS 10 ... Walking movement evaluation apparatus, 11 ... Hip joint information acquisition part, 12 ... Knee joint information acquisition part, 13 ... Ankle joint information acquisition part, 14 ... Walking speed detection part, 15 ... Height detection part, 21 ... Walking movement variable acquisition part , 22: walking speed / height adjustment unit, 23: regression analysis unit, 24: walking movement age index acquisition unit, 25: display data creation unit, 26: storage unit

Claims (8)

A walking motion variable acquisition unit configured to obtain at least a joint angle and an angular velocity of a lower limb during walking of a person to be evaluated as a walking motion variable;
An adjusting unit that adjusts the walking motion variable acquired by the walking motion variable acquiring unit based on the walking speed of the person to be evaluated;
A regression analysis unit that performs regression analysis using the walking motion variable adjusted by the adjustment unit as an explanatory variable and age as a target variable;
A gait movement evaluation device comprising: an age index acquisition unit for acquiring an index of the aging degree of the walking motion of the person to be evaluated from the result of the regression analysis by the regression analysis unit. The walking motion evaluation device according to claim 1, wherein the walking motion variable acquisition unit acquires, as a walking motion variable, an angle and an angular velocity including at least the ankle joint among a hip joint which is a joint of the lower leg, a knee joint, and an ankle joint. . Furthermore, the walking motion variable acquisition unit acquires, as a walking motion variable, a torque and torque power including at least the ankle joint among a hip joint, a knee joint, and an ankle joint, which are joints of the lower leg.
The walking motion evaluation device according to claim 1, wherein the regression analysis unit adds torque and torque power as explanatory variables. The walking movement evaluation device according to any one of claims 1 to 3, which displays an index of the degree of age acquired by the degree of age index acquiring unit. The walking motion evaluation device according to claim 4, further comprising displaying predetermined peaks corresponding to each walking motion variable as a regression analysis result in the regression analysis unit as a profile of the walking motion aging degree. The walking operation evaluation apparatus according to any one of claims 1 to 5, wherein the adjusting unit also adjusts a walking operation variable based on a height of the evaluation target person. Walking motion variable acquisition processing for acquiring at least the angle and angular velocity of the joints of the lower limbs of the person to be evaluated during walking as the walking motion variable;
An adjustment process of adjusting the walking motion variable acquired by the walking motion variable acquisition process based on the walking speed of the person to be evaluated;
Regression analysis processing for performing regression analysis using the walking motion variable adjusted by the adjustment processing as an explanatory variable and age as a target variable;
A walking motion evaluation method, comprising: an aging degree index acquisition process for acquiring an index of the aging degree of the walking motion of the person to be evaluated from the result of regression analysis by the regression analysis process. A walking motion variable acquisition procedure for acquiring at least the angle and angular velocity of the joints of the lower limbs at the time of walking of the evaluation subject as walking motion variables,
An adjustment procedure for adjusting the walking motion variable acquired by the walking motion variable acquisition procedure based on the walking speed of the person to be evaluated;
A regression analysis procedure for performing regression analysis with the walking motion variable adjusted by the adjustment procedure as an explanatory variable and age as a target variable;
An age index acquisition procedure for acquiring an index of the aging degree of the walking motion of the person to be evaluated from the result of regression analysis according to the regression analysis procedure;
A program that causes a computer to execute.

Images