JP2007125368A - Walking analyzer and walking analyzing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a walking analyzer and a walking analyzing method by which walking ability of a user can be universally estimated without placing an excess load on the user. <P>SOLUTION: Front-rear acceleration, right-left acceleration, and up-down acceleration of the waist part of a user are detected by a front-rear acceleration detecting part 12, a right-left acceleration detecting part 14, and an up-down acceleration detecting part 16 of an accelerometer 10 worn on the waist part of the user. A preliminarily obtained relation between an estimated indicator in a specific period and the walking ability is stored in a ROM 26. Walking ability of the user is derived by a CPU 24 on the basis of the changes of the acceleration in time detected by the accelerometer 10 and the relation concerning walking ability stored in the ROM 26. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a walking analysis apparatus and method for analyzing a user's walking ability.

  Examples of such walking analysis apparatus or walking analysis method include the following. First, the technology disclosed in the following Patent Document 1 detects acceleration of a user's waist with an acceleration sensor, detects an acceleration peak by differentiating an output signal from the acceleration sensor, and the peak value Is determined to be abnormal walking (stumbling) when a predetermined threshold is exceeded.

  In addition, the technique disclosed in Patent Document 2 described below is based on the fact that an acceleration sensor is attached to the user's waist, and the user's walking state is determined from the angular acceleration around the horizontal axis detected by the acceleration sensor. It is a gait analysis device that can determine whether or not it is a walking state. Specifically, the angular acceleration output signal from the sensor is frequency-analyzed, and it is determined whether or not the foot is walking by the peak level value that appears at twice the walking frequency. This is because, in normal walking, the same level of rotation occurs around the left and right horizontal axes of the waist with the “stepping” and “kicking up” movements. A large peak at a frequency (half step) that is twice the walking frequency (for one step), but when you walk on a sliding foot, the “pick up backward” action becomes weak, so the frequency is twice the walking frequency. It takes advantage of the fact that the peak level of the lowering.

In addition, the technique disclosed in Patent Literature 3 below attaches an acceleration sensor to the waist, calculates the difference between the peak and valley peak of the vertical acceleration component and the difference between the peak and valley peak of the traveling direction acceleration component, The walking speed and the stride are estimated as the walking ability by the relational expression between the prepared difference and the walking speed. A marker is attached to the body, and the trajectory of the marker is measured with a camera. A coordinate axis is set in the measurement space in advance, and the walking speed and stride are estimated using the movement position and movement time of the marker.
JP-A-10-165395 JP 2000-006608 A JP 2005-114537 A

  However, according to the technique disclosed in Patent Document 1, it is sufficient to detect an acceleration peak and determine that the peak is a stumbling, but it does not detect or estimate the user's walking ability itself. .

  Moreover, in the technique of Patent Document 2, since the rotation of the waist depends on not only the lower limbs but also the movements of the upper limbs, even if the peak level of the frequency twice the walking frequency is relatively low, Is not directly related to walking ability. Moreover, although the fall of the whole leg can be detected by detecting sliding foot walking, the walking ability itself is not detected or estimated as in the case of the cited document 1.

  Moreover, according to the technique disclosed in Patent Document 3, the walking speed or the stride can be estimated as the walking ability. However, the paragraph numbers [0025] and [0026] of Patent Document 3 describe that there is a correlation between the difference between the peak and valley peaks of each acceleration component and the walking speed, as shown in FIG. As shown, it is a result obtained by measuring a specific subject (two people), and cannot be applied to everyone. Also, with this method of simply detecting the vertical acceleration or traveling acceleration peak, the walking ability of each foot cannot be determined individually. If the difference in walking ability between the left and right cannot be clarified, it is insufficient for quantifying walking ability.

  This invention is made | formed in view of the above present conditions, Comprising: It aims at providing the walk analysis apparatus and the walk analysis method which can estimate walking ability more universally.

  As a result of intensive studies by the present inventors, in a period in which the user is performing a specific walking motion, the time change of the acceleration in the vertical direction, the front-rear direction, and the horizontal direction of the waist during that period and the walking ability We found that there is a correlation between them.

  That is, in order to solve the above problems, the walking analysis apparatus of the present invention includes a vertical acceleration that is an acceleration in the vertical direction of the waist during walking, a longitudinal acceleration that is an acceleration in the longitudinal direction of the waist during walking, and a walking Specific acceleration at the time of walking based on the time change of at least one of the accelerations, and an acceleration measuring means for detecting the left and right accelerations which are accelerations in the left and right directions of the waist in A period extracting means for extracting a specific period during which an action is performed, and an estimated index calculating means for calculating an estimated index related to walking ability during walking based on a time change in the specific period of at least one of the accelerations And the estimated index calculated by the estimated index calculating means and the relationship between the estimated index prepared in advance and the walking ability. It characterized by having a a walking ability estimating means for estimating a walking ability with.

  Further, the walking analysis method of the present invention is based on vertical acceleration, which is acceleration in the vertical direction of the waist during walking, longitudinal acceleration, which is acceleration in the longitudinal direction of the waist during walking, and acceleration in the lateral direction of the waist during walking. Based on an acceleration measurement step for detecting a certain lateral acceleration and measuring the temporal change thereof, and extracting a specific period during which a specific walking motion is performed based on the temporal change of at least one of the accelerations. A period extracting step, an estimated index calculating step for calculating an estimated index related to walking ability during walking based on a time change of at least one of the accelerations in the specific period, and the estimated index calculating means Using the calculated estimated index and the relationship between the estimated index prepared in advance and the walking ability And having a walking ability estimating step of estimating a row capability, the.

  As a means for detecting each acceleration of the waist, a form in which an accelerometer as an acceleration measuring means is attached to the waist of the user can be exemplified. In this case, the walking ability can be estimated particularly easily.

  In the gait analysis device or the gait analysis method of the present invention, the period extracting means extracts two time points from the time points when a specific walking motion is performed based on the time change of the acceleration, and these two time points are extracted. It is preferable to extract between two time points as the specific period.

  Further, the period extracting unit extracts the specific period in which a specific walking motion is performed on each of the left and right feet, and the walking ability estimating unit may estimate the walking ability on the left and right feet. .

  As an embodiment of the walking analysis device of the present invention, a means for converting walking ability into a specific identification symbol may be provided, and a display unit for displaying the identification symbol may be provided. The identification symbol is a numerical value indicating walking ability, indicating the level of walking ability step by step (large, medium, small, level, etc.), indicating whether or not a certain level of walking ability has been cleared (pass or fail, safe or Out), or a combination thereof. Further, it may be provided with means for determining whether or not the strength of the estimated walking ability is below the reference, and if it is below the reference, a means for teaching a measure that leads to improvement of the walking ability may be provided.

  Here, examples of the walking ability include at least one selected from dorsiflexion force, knee extension force, walking speed, and stride length. In addition to the above, the walking ability includes, in addition to the above, the step distance (the distance between the left and right feet in the left-right direction), the movable range angle of each joint (hip joint, knee joint, ankle joint), the torque of each joint, the extension of each joint, Examples include flexor and extensor strength, floor reaction force, and the like. In addition, the left / right balance of each walking ability, the distortion of the lower limb skeleton, and the like can be exemplified.

  In addition, it has information storage means for storing walking ability information determined based on three-dimensional acceleration using the walking analysis device, and point calculation means for calculating points based on walking ability information. A health maintenance and promotion system may be configured.

  According to the gait analysis apparatus or the gait analysis method of the present invention as described above, since the temporal change in the acceleration of the waist during a specific period is correlated with the walking ability, the walking ability is estimated only by detecting the acceleration of the waist. can do. In addition, a specific period in which a specific walking motion is performed is extracted based on the temporal change of each acceleration of the waist that is output based on the walking motion, and the walking ability is calculated from an estimated index based on the temporal change of each acceleration during the specific period. Since the estimation is performed, the specific period can be reliably extracted even if the timing at which the specific walking motion is performed by the user varies, and thus the walking ability can be estimated more universally.

  Furthermore, it is possible to easily extract the specific period in which the specific walking motion is performed by detecting at least two time points at which the specific walking motion is performed and extracting the time period as the specific period. Furthermore, by extracting the specific period during which a specific walking motion is performed on each left and right foot and estimating the walking speed on each left and right foot, when used for the purpose of rehabilitation, etc., the effects of rehabilitation and the need for rehabilitation The site and the like can be clarified and efficient treatment can be performed.

  Further, the walking ability determined based on the three-dimensional acceleration cannot be tampered with by shaking the device for measuring the ability, so that information storage means for storing such information, and points based on the information are obtained. The reliability of the health maintenance / promotion system having the point calculation means for calculating is remarkably increased.

  Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. FIG. 1 is a block diagram showing an example of the walking analysis apparatus of the present invention. An apparatus 100 shown in FIG. 1 detects an acceleration of a user's waist during walking and outputs an accelerometer 10 as an acceleration measuring unit that outputs a signal corresponding to the acceleration (hereinafter referred to as an acceleration signal), and an accelerometer. And an arithmetic unit 20 that receives the acceleration signal from the terminal 10 and estimates the walking ability of the user. Furthermore, a time measuring unit 30 that measures walking time, a display unit 40 that can display information such as results of estimating walking ability, user information, etc., a result of determining the strength of walking ability, etc. And a recording unit 50 that can store user information and the like. The accelerometer 10, the calculation unit 20, the time measurement unit 30, the display unit 40, and the recording unit 50 can be integrated. For example, the accelerometer 10, the calculation unit 20, the time measurement unit 30, the display unit 40, and the recording unit 50 can be integrated with each other. . For example, a thing provided with a hook that can be suspended on a belt, trousers, skirt or the like, such as a pedometer.

  The accelerometer 10 includes an X-direction acceleration detection unit 12 as a longitudinal acceleration detection unit that detects the longitudinal acceleration of the waist during walking of the user, and a Y-direction acceleration detection unit as a lateral acceleration detection unit that detects the lateral acceleration of the waist. 14 and a Z-direction acceleration detection unit 16 as a vertical acceleration detection unit that detects the vertical acceleration of the waist. Each of the detection units is integrated into an accelerometer 10 so that if the accelerometer 10 is attached to the user's waist, all longitudinal acceleration, lateral acceleration, and vertical acceleration can be detected. It has become. The acceleration in each direction detected by each of these detection units is an electrical signal corresponding to each acceleration (respectively a longitudinal acceleration signal, a lateral acceleration signal, and a vertical acceleration signal, respectively), and is independently an arithmetic unit. 20 is output.

  As the accelerometer 10, a generally known acceleration sensor can be used. For example, a triaxial acceleration sensor using a piezoelectric element, a capacitance type triaxial acceleration sensor, or the like can be used. In the case of a triaxial acceleration sensor, the longitudinal acceleration detection unit 12, the lateral acceleration detection unit 14, and the vertical acceleration detection unit 16 can be a single detection element. Alternatively, the accelerometer 10 may be used in combination with a uniaxial or biaxial acceleration sensor.

  The calculation unit 20 includes an A / D converter 22, a CPU 24 as a calculation device, a ROM 26 as a storage device, and a RAM 28. The A / D converter 22 converts the signal from the accelerometer 10 into a digital signal, and the acceleration signal digitized from the A / D converter 22 is transmitted to the CPU 24, the ROM 26, and the RAM 28, respectively. It has become. Digitized signals (longitudinal acceleration signal, lateral acceleration signal, and vertical acceleration signal) are temporarily stored in the RAM 28 and subjected to predetermined processing by the CPU 24. For example, the RAM 28 stores a time-varying waveform of the waist acceleration signal together with the time information from the time measuring unit 30. As the time-varying waveform of the acceleration signal, for example, several cycles of walking motion can be stored in the RAM 28.

  In addition, the ROM 26 stores a program for extracting a timing and a period (specific period) for performing a specific walking motion from the longitudinal acceleration signal, the lateral acceleration signal, and the vertical acceleration signal stored in the RAM 28. As shown in FIG. 7, the specific walking movement refers to a heel-contacting action, a sole-grounding action, a toe-off action, a mid-stance phase, and the like. Here, the heel contact operation is an operation in which the heel of one foot is in contact with the ground, the sole contact operation is an operation in which the entire sole of one foot is in contact with the ground, and the toe-off operation is performed on the other foot. This is the action of the toes taking off.

  This program is executed by the CPU 24, and this program, the ROM 26 for storing the program, and the CPU 24 constitute a period extracting means in this embodiment. Further, the ROM 26 stores a program for calculating an estimated index based on each acceleration within a period determined to be a specific period by the period extracting means from a time change of the acceleration signal stored in the RAM 28. . Similar to the period extracting means, the program and the CPU 24 constitute an estimated index calculating means in the present embodiment. The ROM 26 stores a relationship (for example, a relational expression) between an estimated index prepared in advance and walking ability. The ROM 26 stores a program for estimating walking ability from this relationship and each estimated index calculated by the estimated index calculating means. The program and the CPU 24 constitute walking ability estimation means in this embodiment.

  In addition, the gait analysis apparatus of the present embodiment may have an input / output interface that can exchange data with the outside. In addition, the walking analysis apparatus according to the present embodiment may include an input unit that inputs user information and the like. Moreover, you may make it provide the output part which outputs a result to a user.

  First Embodiment Subsequently, an embodiment of the present invention will be described in more detail. In the following embodiments, a walking analysis device will be described, but it also serves as a description of a walking analysis method. First, as a first embodiment, an embodiment in which dorsiflexion force is estimated as walking ability will be described.

  As a result of intensive studies by the present inventors, during a specific period in which a specific walking motion is performed during walking, between the specific estimated index calculated based on the temporal change of the waist acceleration and the strength of the dorsiflexion force Were found to be correlated. That is, the walking analysis apparatus 100 according to the first embodiment includes acceleration measuring means (accelerometer 10) that detects accelerations in the vertical direction, the front-rear direction, and the horizontal direction at the waist during walking, and at least one of the accelerations. Period extracting means (ROM 26, CPU 24) for extracting a specific period during which a specific walking motion related to the strength of dorsiflexion force is performed based on the time change of two accelerations, and the specification of at least one of the accelerations And an estimated index calculating means (ROM 26, CPU 24) for calculating an estimated index related to the strength of the dorsiflexion force based on the time change in the period. Furthermore, it has walking ability estimation means (ROM 26, CPU 24) for estimating the dorsiflexion force using the estimated index calculated by the estimation index calculation means and the relationship between the estimated index prepared in advance and the strength of the dorsiflexion force. Here, the walking ability estimation means may include storage means (storage device, RAM 28, ROM 26) in which the relationship between the estimated index prepared in advance and the strength of the dorsiflexion force is stored.

  In a more specific configuration of the first embodiment, the estimation index may be an average longitudinal acceleration during a specific period. In this case, the strength of the dorsiflexion force can be estimated by obtaining in advance the relationship between the average longitudinal acceleration and the strength of the dorsiflexion force. Here, the average acceleration means a value obtained by integrating the time change of acceleration in the period by time and dividing the value by the period.

  According to the walking analysis apparatus of the first embodiment as described above, the relationship between the acceleration of the waist during walking and the strength of the dorsiflexion force of the human body (hereinafter referred to as the acceleration-dorsiflexion force relationship) is stored in advance. Since it is stored in the device, if the acceleration of the waist during walking is detected by the acceleration measuring means, it is possible to determine the strength of the user's dorsiflexion force by the arithmetic device from the acceleration-dorsiflexion force relationship. it can.

  The correlation between the strength of the dorsiflexion force and the acceleration of the waist is thought to be due to the following reason. In other words, the dorsiflexion is performed by the shin muscle (anterior tibialis anterior muscle) and also uses the acceleration force generated by the backward kicking force. When the kicking force is weak, the dorsiflexion force decreases and the toes easily fall. It is obvious that the kicking action has a function of generating a driving force for walking as well as a function of assisting the dorsiflexion action. Therefore, it is considered that there is a correlation between the acceleration of the lumbar region and the strength of the dorsiflexion force related to the propulsive force of walking, and if the acceleration of the lumbar region is detected, the strength of the dorsiflexion force can be derived. Is possible.

  Further, according to the kicking-out operation, the acceleration of the waist is mainly directed in the front-rear direction, so that there is a further correlation between the back-and-forth acceleration of the waist and the strength of the dorsiflexion force. Therefore, the relationship between the longitudinal acceleration of the lumbar region and the strength of the dorsiflexion force is obtained in advance as an acceleration-dorsiflexion force relationship, the longitudinal acceleration of the lumbar region is detected, and the average longitudinal acceleration in a specific period is used as an estimation index, the acceleration− The strength of the user's dorsiflexion force can be determined more accurately from the dorsiflexion force relationship.

  In the first embodiment, since the strength of the dorsiflexion force can be estimated, for example, it is possible to detect a lowering of the lower limbs that induces “stumbling” which is a main factor of falling. This is due to the following reason.

  First, walking refers to alternately swinging left and right feet forward. A leg that touches the ground and supports weight is called a stance leg, and a leg that swings forward from the ground is called a free leg. In walking, each of the left and right feet has a stance phase in which the feet are on the ground and a swing phase in which the feet are separated from the ground. During walking, the period in which the left and right feet are in the stance phase simultaneously is the period for supporting both legs, and the period in which only one leg is in the stance phase is the period for supporting the single leg.

  The stance phase starts with the heel of the foot that has become a free leg in contact with the ground (heel contact), and the bottom of the foot is in contact with the ground surface by touching the toe side to the ground ( (Foot grounding), the state where the bottom of the foot is in contact with the floor surface, the state where the heel part is separated from the floor surface (the heel-off area), and the toes (tips) are separated from the floor surface, so that the foot is on the floor surface. End in a state of leaving (a point away from the toes). Accordingly, in each foot, the stance phase is from the heel contact to the toe separation, and the swing phase is from the toe contact to the heel contact.

  If the dorsiflexion force decreases, the toe hangs down during the swing phase (a period in which the foot is swung forward) during one cycle of walking movement (one step left and right), so that a sufficient space between the ground and the toes can be secured. Can not. Therefore, it is easy to trip and toppling over. Therefore, by estimating the strength of the dorsiflexion force, it is possible to detect “easy to trip” that directly affects the ease of falling. Thereby, a fall can be prevented at an early stage. In addition, since a trip is a cause of a fall, detecting the trip itself does not prevent a fall as an actual problem, but according to the walking analysis device of the present invention, the point is Since “easy to trip” is discriminated, it is possible to more effectively take measures to prevent falls.

  In the gait analysis device of the first embodiment, a means for converting the strength of the dorsiflexion force into a specific identification symbol may be provided, and a display unit for displaying the identification symbol may be provided. The identification symbol is a numerical value indicating the dorsiflexion force, indicating the level of dorsiflexion force in stages (large, medium, small, level, etc.), indicating whether or not a certain level of dorsiflexion force has been cleared (pass or fail) , Safe or out), or a combination thereof. Further, there may be provided means for determining whether or not the strength of the determined dorsiflexion force is equal to or less than a reference, and if it is equal to or less than the reference, a means for teaching a measure that leads to fall prevention may be provided.

  Furthermore, according to the walking analysis apparatus of the first embodiment, since it is possible to detect a specific decline in the lower limbs, such as a decrease in dorsiflexion force, muscle strength related to the strength of dorsiflexion force (for example, front It is possible to specifically find out that the tibial muscle, etc.) is declining. Therefore, when teaching users measures to prevent falls, teach the user more specifically exercises to increase muscles that are directly related to dorsiflexion, rather than exercises that involve the entire lower limbs. can do. Therefore, it is more effective in preventing falls.

  Although it is possible to grasp the decrease in dorsiflexion force by consulting with a specialist such as a physical therapist or walking analysis by motion capture etc., it is troublesome to consult with a specialist in these methods. Or require large-scale equipment. In the present invention, since it is only necessary to detect the acceleration of the lumbar region, for example, it is only necessary to wear the lumbar region and walk, and a decrease in dorsiflexion force can be easily detected.

  In the more specific configuration of the first embodiment, the specific period is a period from the time when the kicking action of kicking with one foot is performed to the time when the middle stage of standing with only the other leg is performed. Can be illustrated. The propulsive force generated by the kicking motion during walking most appears in the longitudinal acceleration of the waist during the period from the kicking motion of kicking with one foot to the middle of the stance when standing with only the other foot. Therefore, if the period during which this walking motion is performed is a specific period, there is a further correlation between the longitudinal acceleration of the waist and the strength of the dorsiflexion force during this specific period. Therefore, if the relationship between the lumbar longitudinal acceleration and the strength of the dorsiflexion force in this specific period is obtained in advance, the dorsiflexion force can be estimated by calculating the average longitudinal acceleration in the specific period as an estimation index. .

  Further, as the specific period, a period from the time when the heel-grounding operation in which the heel of one foot is grounded to the time when the sole-grounding operation in which the entire sole of the one foot is grounded can be exemplified. . Further, as the specific period, a period from the time when the heel-contacting operation in which the heel of one foot is grounded to the time of performing the toe-off operation in which the toe of the other foot is detached can be exemplified. .

  In addition, the strength of the dorsiflexion force may be an index expressed as a ratio of a change in ankle dorsiflexion angle during the specific period. In this case, the strength of the dorsiflexion force can be estimated by obtaining in advance a relationship between the estimated index in the specific period and the rate of change of the dorsiflexion angle. The dorsiflexion angle is an angle formed by a direction from the ankle joint toward the shin and a direction from the ankle joint toward the toe.

  As shown in FIG. 3, the dorsiflexion angle of the right foot with respect to one foot (hereinafter, described as the right foot) is minimum during the both-leg support period in which both the right foot and the left foot are in contact with the ground during the walking motion. Specifically, the dorsiflexion angle of the right foot is usually the smallest during the walking motion in the left foot contact where the entire bottom of the left foot is in contact with the ground. More specifically, when the left sole is in contact with the ground, where the entire bottom of the left foot is in contact with the ground, the weight gradually shifts from the left footpad to the toe, but the dorsiflexion angle of the right foot is the middle of this forward weight shift. It becomes the minimum at a certain time point (time point O) from the time point to the toe departure point where the toe leaves the ground.

  In addition, the dorsiflexion angle of the right foot is normally maximized at the time immediately after the right toe take-off operation shown in FIG. Specifically, the dorsiflexion angle of the right foot is maximized at a time point (time A) before and after the operation of swinging the right foot forward after the right foot is separated from the ground by kicking the right foot. It is normal. As shown in FIG. 6, when the dorsiflexion force is reduced after the kicking-out operation (subject 60s), the toe hangs down and the dorsiflexion angle remains large, whereas the dorsiflexion force Is sufficiently maintained (subject in the 20s), the dorsiflexion angle becomes smaller again after dorsiflexion.

  Therefore, when the dorsiflexion force is strong, the dorsiflexion at the point A with respect to the difference (β−α) between the dorsiflexion angle at the time O (hereinafter referred to as α) and the dorsiflexion angle at the time A (hereinafter referred to as β). The ratio of the difference (β−γ) between the angle (β) and the dorsiflexion angle (hereinafter referred to as γ) in the period from kicking to starboard contact (see FIG. 3) is relatively large. . That is, when the dorsiflexion force is strong, the value represented by (β−γ) / (β−α) is large. On the other hand, when the dorsiflexion force is reduced, the ratio of β-γ to β-α is relatively small, and the value represented by (β-γ) / (β-α) is small. It becomes.

  That is, the strength of the dorsiflexion force can be evaluated based on the magnitude of (β−γ) / (β−α) (Expression 1). Therefore, the acceleration-dorsiflexion force relationship can be obtained by obtaining in advance the relationship between the waist acceleration and the strength of the dorsiflexion force expressed by Equation 1. Specifically, the dorsiflexion obtained by measuring the waist acceleration (especially the longitudinal acceleration between the kicking motion and mid-stance) and the dorsiflexion angle during walking for multiple subjects. By calculating the strength of the dorsiflexion force defined by Equation 1 from the angle and obtaining the relationship between the strength of the dorsiflexion force and the acceleration of the waist as a function, for example, the above-mentioned acceleration-dorsiflexion force relationship is obtained. be able to.

  In the first embodiment, the ROM 26 stores an acceleration-dorsiflexion force relationship obtained in advance. The acceleration-dorsiflexion force relationship stored in the ROM 26 is, for example, as shown in FIG. The acceleration-dorsiflexion force relationship is represented by the straight lines in the graphs 1 to 3 shown in FIG.

  All acceleration-dorsiflexion force relations are the average acceleration of longitudinal acceleration (hereinafter simply referred to as average longitudinal acceleration) during a specific period from the kicking motion to the middle left stance (see Fig. 3), and the strength of the dorsiflexion force. The relationship is defined. In this embodiment, the strength of the dorsiflexion force is defined by the above-described (β−γ) / (β−α) (Equation 1). Specifically, as shown in FIG. 4, the dorsiflexion angle is the maximum when α is the dorsiflexion angle at the point O when the dorsiflexion angle is minimum (corresponding to the vicinity of the ground contact with the left sole in the period of supporting both legs in FIG. 3). Β is the dorsiflexion angle at time A (corresponding to before and after the swinging motion in FIG. 3), and substantially corresponds to the motion from the swinging leg phase to the starboard grounding shown in FIG. ) Is the dorsiflexion angle. As the dorsiflexion angle γ, dorsiflexion angles γ1, γ2, and γ3 at time points B, C, and D, which will be described later, can be employed. In each graph of FIG. 2, the vertical axis represents the strength of the dorsiflexion force expressed by the above formula 1, and the horizontal axis represents the average acceleration.

  Moreover, the graph 1 of FIG. 2 shows the relationship between the strength of the mid-stance dorsiflexion force immediately after the kicking-out operation and the average acceleration. Here, the strength of the mid-stance dorsiflexion force is the strength of the dorsiflexion force in the mid-left stance phase shown in FIG. 3 (time B: mid-right stance when the kicking action is the left foot). This corresponds to the case where the dorsiflexion angle γ1 at the time B shown in FIG. 4 (corresponding to the middle stage of the left stance in FIG. 3) is adopted as the angle γ. That is, using the acceleration-dorsiflexion force relationship shown in the graph 1 of FIG. 2, it is possible to determine the strength of the mid-stance dorsiflexion force from the average acceleration.

  On the other hand, the graph 2 in FIG. 2 shows the relationship between the minimum point dorsiflexion strength and the average acceleration when the dorsiflexion angle is minimized in the operation from the kicking-out operation to the starboard contact. Here, the strength of the minimum point dorsiflexion force is the strength of the dorsiflexion force at the time when the dorsiflexion angle becomes minimum in the right leg swing leg phase shown in FIG. This corresponds to the case where the dorsiflexion angle γ2 at the time is adopted. That is, when this acceleration-dorsiflexion force relationship is used, the strength of the minimum point dorsiflexion force in the operation over the right leg swing leg phase and starboard landing can be estimated from the detected average acceleration.

  Graph 3 in FIG. 2 shows the relationship between the strength of the maximum point dorsiflexion force and the average acceleration. Here, the strength of the maximum point dorsiflexion force is the time when the dorsiflexion angle becomes the maximum again (or right) in the right leg swing leg period including time point A after the time point A in the walking motion shown in FIG. This shows the strength of the dorsiflexion force at the time of heel contact), and corresponds to the case where the dorsiflexion angle γ3 at time D shown in FIG. That is, when the acceleration-dorsiflexion force relationship shown in the graph 3 of FIG. 2 is used, the strength of the maximum point dorsiflexion force can be estimated from the average acceleration.

  These acceleration-dorsiflexion force relationships are measured for a plurality of subjects by measuring the acceleration of the waist during walking, and simultaneously measuring the change in dorsiflexion angle as shown in FIG. The strength of the bending force is calculated, and the relationship between the acceleration and the strength of the dorsiflexion force is obtained from a plurality of data (corresponding to the dots in FIG. 2) that are pairs of the acceleration and the strength of the dorsiflexion force (in FIG. 2). (Corresponding to a straight line) can be obtained, for example, by deriving by the least square method or the like.

  In FIG. 2, the relationship between the strength of the maximum point dorsiflexion force and the average acceleration seems to be the clearest. Therefore, in the acceleration-dorsiflexion force relationship in the present embodiment, the strength of the maximum point dorsiflexion force can be typically handled as the strength of the dorsiflexion force after the kicking-out operation. Note that when an index other than the present embodiment (an index other than that expressed by Equation 1) is adopted as the strength of the dorsiflexion force, the dorsiflexion force at a time other than the D time point when the dorsiflexion angle is maximized. It is also assumed that there is the most relationship between the average acceleration. In that case, the strength of the dorsiflexion force at other time points can be adopted as a representative index.

  Hereinafter, the operation and operation of the walking analysis apparatus 100 of the first embodiment will be described. First, when the walking analysis device 100 of the present embodiment is mounted on the waist and walking is started, the accelerometer 10 of the walking analysis device 100 measures temporal changes in the waist longitudinal acceleration, lateral acceleration, and vertical acceleration. The accelerometer 10 outputs the detected time change of acceleration as a waveform of an electric signal (acceleration signal). The acceleration signal output from the accelerometer 10 is digitized by the A / D converter 22 and temporarily stored in the RAM 28. FIG. 5 shows an example of a time change of an acceleration signal (longitudinal acceleration signal, lateral acceleration signal, vertical acceleration signal) detected by the accelerometer 10. It can be seen that each acceleration changes with time. In FIG. 5, the longitudinal acceleration is shown as positive on the forward side, the lateral acceleration is shown as positive on the right side, and the vertical acceleration is shown as positive on the upward side. Each time change of the acceleration corresponds to a specific walking motion of the walking motion.

  When an acceleration signal (longitudinal acceleration, lateral acceleration, vertical acceleration) corresponding to one cycle of walking motion is stored in the RAM 28, a specific period is extracted from the time change of the acceleration signal by the period extracting means constituted by the CPU 24 and ROM 26. The timing and period (specific period) when the walking motion is performed are extracted. Specifically, it is possible to detect the peak of the acceleration signal by differentiating the acceleration signal with respect to time and determine that a specific walking motion starts or ends at the peak. For example, in FIG. 5, it can be determined that the kick-out operation of FIG. 3 is performed at the time when the longitudinal acceleration and the vertical acceleration and the lateral acceleration are maximized (time X) while the longitudinal acceleration is minimized. Then, after the kicking operation, the time point (Z time point) when the vertical acceleration and the horizontal acceleration become minimum can be determined as the right toe separation (when the kicking operation is the right foot) in FIG. Further, after the toe-off operation, the longitudinal acceleration and the lateral acceleration gradually increase, but the vertical acceleration gradually decreases and then becomes minimal. The time point at which the vertical acceleration is minimized (Y time point) can be determined as the middle right stance phase (when the kicking motion is the right foot).

  As described above, if the time point when the specific walking motion is performed can be grasped, the specific period during which the specific walking motion is performed is extracted from at least two time points among them. Then, the average longitudinal acceleration is calculated by the CPU 24 based on the time change of the acceleration during the specific period. Specifically, an average acceleration in the front-rear direction is calculated as an estimated index from a time change of the front-rear acceleration in a specific period between the time point determined as the kicking motion and the time point determined as the middle stance.

  Using the average acceleration calculated in this manner as an estimation index, the strength of the dorsiflexion force is estimated from the acceleration-dorsiflexion relationship. Specifically, the CPU 24 applies the calculated average acceleration to the acceleration-dorsiflexion force relationship as shown in FIG. 2 stored in the ROM 26 and stores the strength of the dorsiflexion force corresponding to the average acceleration. Can be derived. Here, the strength of the dorsiflexion force is expressed by Formula 1. Furthermore, as the strength of the dorsiflexion force, the strength of the mid-stand stance dorsiflexion force, the strength of the minimum point dorsiflexion force, and the strength of the maximum point dorsiflexion force shown in FIG. And stored in the RAM 28. Alternatively, the calculation result may be automatically stored as data in the recording unit 50.

  Further, the current walking age of the user can be calculated from the estimated strength of dorsiflexion. Specifically, the ROM 26 stores in advance the relationship between the strength of the dorsiflexion force and the walking age, and the calculated strength of the dorsiflexion force is related to the relationship between the strength of the dorsiflexion force and the walking age. By applying, the walking age can be calculated. The calculated walking age is temporarily stored in the RAM 28. Alternatively, the walking age determination result may be automatically stored in the recording unit 50.

  As described above, when the CPU 24 derives the strength of the dorsiflexion force represented by Equation 1, the result can be displayed on the display unit 40. In this case, the user selects the item to be displayed based on the strength of the middle stance dorsiflexion force, the strength of the minimum point dorsiflexion force, the strength of the maximal point dorsiflexion force, and the estimated result of the selected item is displayed. can do. Alternatively, all estimation results may be automatically displayed.

  Furthermore, the walking age based on the estimated strength of the dorsiflexion force can be displayed. Further, by displaying an exercise method for preventing falls according to the age of walking, the user can be instructed with an exercise guideline.

  As described above, according to the walking analysis apparatus 100 of the present embodiment, the strength of the user's dorsiflexion force can be estimated as the walking ability, so that the “trigger” that is the main factor of the fall is induced. It is possible to more easily detect the deterioration of the lower limb. Furthermore, since it is possible not to detect the trip itself but to directly detect a decrease in dorsiflexion force that induces the trip, it is possible to take measures to prevent falls earlier.

  (2nd Embodiment) Next, the walk analysis apparatus 100 concerning 2nd Embodiment is demonstrated. The walking analysis apparatus 100 according to the second embodiment estimates lower limb muscle strength as walking ability. As a result of intensive studies, the present inventors have determined that during a specific period when a specific walking motion is performed, between a specific estimated index calculated based on a temporal change in acceleration of the lower back and lower limb muscle strength, It was found that there was a correlation. That is, the walking analysis apparatus 100 according to the second embodiment includes acceleration measuring means (accelerometer 10) that detects accelerations in the vertical direction, the front-rear direction, and the horizontal direction at the waist during walking, and at least one of the accelerations. Period extracting means (ROM 26, CPU 24) for extracting a specific period during which a specific walking motion related to the lower limb muscle strength is performed based on a time change of two accelerations, and a time of at least one of the accelerations in the specific period And an estimated index calculation means (ROM 26, CPU 24) for calculating an estimated index related to the lower limb muscle strength based on the change. Furthermore, it has walking ability estimating means (ROM 26, CPU 24) for estimating lower limb muscle strength using the estimated index calculated by the estimated index calculating means and the relationship between the estimated index prepared in advance and the lower limb muscle strength. Here, the walking ability estimation means may include storage means (storage device, RAM 28, ROM 26) in which the relationship between the estimated index prepared in advance and the lower limb muscle strength is stored.

  In a more specific configuration in the second embodiment, the estimated index may be an average longitudinal acceleration during a specific period. In this case, the lower limb muscle strength can be estimated by obtaining in advance the relationship between the average longitudinal acceleration and the lower limb muscle strength.

  The reason why the lower limb muscle strength can be estimated by measuring the temporal change of the acceleration of the waist during walking is as follows. That is, the longitudinal acceleration of the waist during walking corresponds to the propulsive force / braking force in the front-rear direction obtained by the lower limb muscle strength. Therefore, there is a further correlation between the lumbar longitudinal acceleration and the lower limb muscle strength. Therefore, the relationship between the longitudinal acceleration of the lower back and the muscle strength of the lower limbs is obtained in advance as the acceleration-lower limb strength relationship, and the user's lower limb strength is measured from the acceleration-lower limb strength relationship using the average longitudinal acceleration during a specific period as an estimation index. can do.

  On the other hand, when walking, kick the leg and go forward. The kicking force of the leg goes in the front-rear direction and goes up and down. Therefore, for example, the strength of the lower limb muscle strength corresponding to the kicking force correlates with the vertical acceleration of the waist during walking. Therefore, the relationship between the vertical acceleration of the lower back and the lower limb muscle strength is obtained in advance as the acceleration-lower limb muscle strength relationship, and the user's lower limb muscle strength is estimated from the acceleration-lower limb muscle strength relationship using the average vertical acceleration during a specific period as an estimation index. can do.

  Furthermore, when measuring the lower limb muscle strength from the longitudinal acceleration or the vertical acceleration, the foot for which the entire sole of the one foot is grounded from the point of time when the heel-grounding operation is performed in which the heel of one foot is grounded during the specific period. The period up to the time when the bottom grounding operation is performed can be used. At the time of heel contact, lower extremity muscle strength (for example, lower extremity muscle strength) that opposes the plantar flexion moment generated around the ankle joint (the moment that rotates the ankle joint in the direction in which the toes touch the ground due to the action from the ground) is required.

  This bottom flexion moment becomes larger as the walking speed at the time of landing on the heel increases. Therefore, when the lower limb muscle strength is small, it is considered that the walking speed at the time of heel-contact is reduced in order to make the plantar flexion moment acceptable. On the other hand, when the muscle strength of the lower limbs is large, a large plantar flexion moment can be tolerated, so it is considered that the walking speed at the time of heel contact is increased. From the above, it is considered that the muscle strength of the lower limbs can be estimated by using the longitudinal acceleration of the lumbar region at the time of heel contact as an estimation index.

  Further, when the heel is touched, the lower limb muscle strength is required to resist the bending moment generated around the knee joint (the moment to rotate the knee joint in the direction in which the knee bends due to the action from the ground). This bending moment increases as the walking speed at the time of landing on the heel increases. Therefore, when the lower limb muscle strength is small, it is considered that the walking speed at the time of the heel-contact is lowered in order to make this bending moment acceptable. On the other hand, when the muscle strength of the lower limbs is large, a large bending moment can be tolerated, so that it is considered that the walking speed at the time of heel contact increases. Therefore, the lower limb muscle strength can be estimated by using the back and forth acceleration of the waist when the heel is touched as an estimation index.

  Further, if the lower limb muscle strength is low, it is difficult to resist the bending moment generated when the heel is touched, so it is considered that there is a strong tendency to perform the heel contact operation while the knee joint is stretched so that the knee does not bend. When the heel contact operation is performed while the knee joint is stretched, the impact from the ground during the heel contact is not absorbed by the flexion of the knee joint. The impact at this time affects the vertical acceleration at the waist. In other words, when the impact is large, the vertical acceleration increases when the upward direction is positive. In other words, when the muscle strength of the lower limbs is low, it is considered that the vertical acceleration when touching the heel increases. On the other hand, when the muscle strength of the lower limbs is high, the bending moment can be countered, so it is considered that there is a strong tendency to absorb the impact by bending the knee joint when the heel is touched. Therefore, when the muscle strength of the lower limbs is high, the impact received from the ground when the heel touches is small, in other words, the vertical acceleration when the heel touches is small. As described above, it is possible to estimate the lower limb muscle strength by detecting the acceleration of the lower back in the heel contact operation.

  Similarly to the above, when measuring muscle strength of the lower limbs from the longitudinal acceleration or the vertical acceleration, the specific period is from the time when the plantar grounding operation is performed in which the entire plantar of one foot is grounded from the time point of the other foot It is possible to set the movement up to the time point when the toe-off operation is performed. The acceleration of the waist during this specific period is related to the braking ability in the walking motion. If the lower limb muscle strength is low, the braking ability naturally decreases. Therefore, the lower limb muscle strength can be measured by detecting the acceleration of the waist during this specific period. More specifically, the longitudinal acceleration of the lumbar region during this specific period is a negative acceleration indicating deceleration, and it can be determined that the greater the absolute value of this negative longitudinal acceleration, the greater the braking ability, that is, the lower limb muscle strength. . Conversely, it can be determined that the smaller the absolute value of the negative longitudinal acceleration, the smaller the braking ability, that is, the lower limb muscle strength.

  Further, as described above, when the lower limb muscle strength is estimated from the longitudinal acceleration or the vertical acceleration, the foot of the other foot starts from the time when the heel contact operation is performed in which the heel of one foot contacts the ground during the specific period. It can be set as the period until the time when the toe-off operation to take off is performed. In this specific period, the walking speed is reduced by the braking force as a whole. Therefore, for the same reason as described above, the lower limb muscle strength can be measured.

  In the configuration of the second embodiment described above, the lower limb muscle strength can be a dorsiflexion force. The dorsiflexion force has a correlation with the acceleration at the waist during walking, and can be easily measured by the apparatus of this embodiment. The correlation between the dorsiflexion force and the acceleration of the waist is considered to be due to the following reason. Walking motion is performed by repeatedly accelerating and decelerating, and one of the muscle strengths used for decelerating is dorsiflexion. Therefore, since the acceleration of the waist during the specific walking motion corresponding to the braking period in the walking motion is affected by the braking force, it is considered that there is a correlation between the acceleration of the waist and the dorsiflexion force. Therefore, it is possible to estimate the dorsiflexion force by using the acceleration of the waist during a specific period as an estimation index.

  Alternatively, the lower limb muscle strength can be a knee extension force. The knee extension force is one of muscle strengths that exert a braking force in walking motion. Therefore, since the acceleration of the waist during a specific period corresponding to the braking period during the walking motion is influenced by the braking force, it is considered that there is a correlation between the acceleration of the waist and the knee extension force. Therefore, it is possible to estimate the knee extension force by using the waist acceleration during the specific period as an estimation index. In this case, the knee extension force can be estimated by calculating the estimated index by obtaining in advance the relationship between the estimated index and the knee extension force based on the time change of each acceleration in the specific period.

  Further, when dorsiflexion force or knee extension force is adopted as the lower limb muscle strength, the foot for which the entire sole of the one foot is grounded from the point of time when the heel-grounding operation for grounding the heel of one foot is performed during the specific period. The period until the time when the bottom grounding operation is performed can be set, and the estimated index can be an average longitudinal acceleration in the specific period. In addition, the specific period is a period from the time when the plantar grounding operation where the entire sole of one foot is grounded to the time when the plantar grounding operation is performed where the foot of the other foot is detached, The estimated index can be an average longitudinal acceleration during the specific period. In addition, the specific period is a period from the time when the heel-grounding operation in which the heel of one foot is grounded to the time when the foot-grounding operation in which the entire sole of the one foot is grounded is performed, and the estimation index is The average vertical acceleration during the period can be used. Further, regarding the dorsiflexion force and knee extension force, it is possible to estimate the lower limb muscle strength more precisely by combining the relationship between the plurality of accelerations and the lower limb muscle strength.

  In the second embodiment, the ROM 26 stores an acceleration-lower limb strength relationship determined in advance. The acceleration-lower limb muscle strength relationship stored in the ROM 26 represents the relationship between the longitudinal acceleration and the strength of the lower limb muscle strength in the present embodiment. The acceleration-lower limb muscle strength relationship stored in the ROM 26 is, for example, as shown in FIGS.

  The relationship between acceleration and lower limb muscle strength is represented by a straight line in FIGS. More specifically, these acceleration-lower limb muscle strength relationships show the relationship between the dorsiflexion force or knee extension force of the lower limb muscle strength and the waist acceleration during walking. These acceleration-lower limb muscle strength relationships can be obtained by obtaining in advance a relationship between longitudinal acceleration and vertical acceleration in each specific walking motion and lower limb muscle strength such as dorsiflexion force and knee extension force. Specifically, for a plurality of subjects, the waist acceleration (longitudinal acceleration, vertical acceleration, etc.) during walking was measured, and at the same time, the lower limb muscle strength such as dorsiflexion force and knee extension force was measured. The relationship between acceleration and lower limb muscle strength (corresponding to the straight line in FIGS. 8 to 13) is derived from a plurality of data (corresponding to dots in FIGS. 8 to 13) paired with muscle strength by, for example, the least square method or the like. Can be obtained.

  Next, the acceleration-lower limb muscle strength relationship shown in FIGS. First, the acceleration-lower limb muscle strength relationship in FIG. 8 shows the relationship between the average longitudinal acceleration and the dorsiflexion force in the specific period A (the period in which the sole contact operation is performed from the heel contact operation) shown in FIG. . When the acceleration-lower limb muscle strength relationship shown in FIG. 8 is used, the user's dorsiflexion force can be measured by calculating the average longitudinal acceleration during the period A.

  The acceleration-lower limb muscle strength relationship in FIG. 9 shows the relationship between the average longitudinal acceleration and the knee extension force in the specific period A shown in FIG. 7 (the period in which the sole contact operation is performed). When the acceleration-lower limb muscle strength relationship shown in FIG. 9 is used, the knee extension force of the user can be measured by calculating the average longitudinal acceleration during the period A.

  The acceleration-lower limb muscle strength relationship in FIG. 10 shows the relationship between the average longitudinal acceleration and the knee extension force in the specific period B shown in FIG. 7 (the period in which the toe-off operation is performed). . When the acceleration-lower limb muscle strength relationship shown in FIG. 10 is used, the knee extension force of the user can be measured by calculating the average longitudinal acceleration during the period B.

  The acceleration-lower limb muscle strength relationship in FIG. 11 shows the relationship between the average longitudinal acceleration and the dorsiflexion force in the specific period of C shown in FIG. 7 (the period in which the toe-off action is performed). When the acceleration-lower limb muscle strength relationship shown in FIG. 11 is used, the user's dorsiflexion force can be measured by calculating the average longitudinal acceleration during the period C.

  The acceleration-lower limb muscle strength relationship in FIG. 12 shows the relationship between the average longitudinal acceleration and the knee extension force in the specific period C (the period in which the toe-offing operation is performed) shown in FIG. When the acceleration-lower limb muscle strength relationship shown in FIG. 12 is used, the knee extension force of the user can be measured by calculating the average longitudinal acceleration during the period C.

  The acceleration-lower limb muscle strength relationship in FIG. 13 shows the relationship between the average vertical acceleration and the knee extension force in the specific period A shown in FIG. 7 (the period in which the sole contact operation is performed). When the acceleration-lower limb muscle strength relationship shown in FIG. 13 is used, the knee extension force of the user can be measured by calculating the average vertical acceleration during the period A.

  The acceleration-lower limb muscle strength relationship includes three variables: an average longitudinal acceleration (Xa) in the period A, an average longitudinal acceleration (Xb) in the period B, and an average vertical acceleration (Xc) in the period A. The acceleration-lower limb muscle strength relationship obtained by multiple regression analysis of the relationship with the dorsiflexion force can be used. These (Xa), (Xb), and (Xc) correspond to estimated indexes. Specifically, the acceleration-lower limb muscle strength relationship in this case is represented by dorsiflexion force (Y1) = A1Xa + B1Xb + C1Xc + D1 (formula a). Furthermore, as a result of collecting the data from a plurality of subjects and performing the multiple regression analysis as described above, the more specific acceleration-lower limb strength relationship in this case is the dorsiflexion force (Y1) = − 0.53Xa−. It is represented by 0.59Xb + 0.1Xc + 0.35 (formula b). The multiple correlation coefficient in the acceleration-lower limb strength relationship is as large as about 0.6. Therefore, by using the acceleration-lower limb muscle strength relationship of the above (formulas a and b) and calculating the variables (Xa), (Xb), and (Xc), respectively, the user's dorsiflexion force can be made more precise. Can be measured.

  The acceleration-lower limb muscle strength relationship includes three variables: an average longitudinal acceleration (Xa) in the period A, an average longitudinal acceleration (Xb) in the period B, and an average vertical acceleration (Xc) in the period A. The acceleration-lower limb muscle strength relationship obtained by performing multiple regression analysis of the relationship with the knee extension force can be used. Specifically, the acceleration-lower limb muscle strength relationship in this case is expressed by knee extension force (Y2) = A2Xa + B2Xb + C2Xc + D2 (formula c). Furthermore, as described above, as a result of collecting the data from a plurality of subjects and performing the above multiple regression analysis, a more specific acceleration-lower limb muscle strength relationship is: knee extension force (Y2) = − 2.29Xa−3.34Xb -0.52Xc-0.5 (formula d). The multiple correlation coefficient in the acceleration-lower limb strength relationship is as large as about 0.74. Therefore, by using the acceleration-lower limb muscle strength relationship of the above (formulas c, d), the variables of (Xa), (Xb), and (Xc) are calculated, respectively, so that the knee extension force of the user can be made more precise. Can be measured.

  Hereinafter, the operation and operation of the walking analysis apparatus 100 according to the second embodiment will be described. First, when the walking analysis apparatus 100 of the present embodiment is mounted on the waist and walking is started, the accelerometer 10 of the walking analysis apparatus 100 detects the longitudinal acceleration, the lateral acceleration, and the vertical acceleration of the waist. The accelerometer 10 outputs the detected time change of acceleration as a waveform of an electric signal (acceleration signal). The acceleration signal output from the accelerometer 10 is digitized by the A / D converter 22 and temporarily stored in the RAM 28. FIG. 7 shows an example of a time change of an acceleration signal (longitudinal acceleration signal, lateral acceleration signal, vertical acceleration signal) detected by the accelerometer 10. It can be seen that each acceleration changes with time. In FIG. 7, the longitudinal acceleration is shown as positive on the forward side, the lateral acceleration is shown as positive on the right side, and the vertical acceleration is shown as positive on the upward side. Each time change of the acceleration corresponds to a specific walking motion of the walking motion as shown in FIG.

  When an acceleration signal (longitudinal acceleration, lateral acceleration, vertical acceleration) corresponding to one cycle of walking motion (may be several cycles) is stored in the RAM 28, the walking motion determination means configured by the CPU 24 and the ROM 26 The timing and period (specific period) at which a specific walking motion is performed are extracted from the time change of the acceleration signal. Specifically, it is possible to detect the peak of the acceleration signal by differentiating the acceleration signal with respect to time and determine that a specific walking motion starts or ends at the peak. Alternatively, focusing on one acceleration selected from longitudinal acceleration, vertical acceleration, and lateral acceleration, it is possible to determine that a specific walking motion starts or ends when the acceleration changes from positive to negative or from negative to positive. it can. Alternatively, it is possible to determine the timing when the acceleration becomes a predetermined ratio with respect to the acceleration at the peak of the acceleration signal as the start time and the end time of the walking motion. Furthermore, among all the period extraction methods described here, the most appropriate one may be selected at any time, or a method in which these period extraction methods are appropriately combined may be employed.

  For example, in the present embodiment, the following determination method is adopted to determine the specific walking motion as shown in FIG. First, the (right) heel contact determination method of FIG. 7 can be performed as follows. (1) A negative peak (P1) having a large longitudinal acceleration is detected, and a minimum point (P2) of the longitudinal acceleration immediately before the large negative peak (P1) is detected. (2) It is determined whether there is a point (P3) where the lateral acceleration changes from positive to negative between the negative peak (P1) and the minimum point (P2). (3) When the point (P3) where the lateral acceleration changes from positive to negative is within the above range, the time point of P3 is determined to be the time point of (right) 踵 ground contact operation. (4) When the point (P3) at which the lateral acceleration changes from positive to negative is not within the above range, the minimum point (P2) is determined as the point of (right) 右 grounding operation. In addition, the above example is a determination in (right) heel contact, and in the case of determining (left) heel contact, in the step (2), the negative peak (P1) and the minimum point (P2) It may be determined whether or not there is a point (P4) between which the lateral acceleration changes from negative to positive.

  Next, after the heel contact is determined as described above, the foot contact can be determined as a point where the vertical acceleration first changes from positive to negative after the heel contact. Further, the point of foot separation can be the first point at which the vertical acceleration after the planting of the sole changes from negative to positive, or the maximum point that appears first, whichever comes first.

  As described above, if the time point when the specific walking motion is performed can be grasped, the CPU 24 calculates the average acceleration during the specific period during which the specific walking motion is performed. Specifically, the average longitudinal acceleration in the period A as an estimation index is obtained by calculating the average acceleration of the longitudinal acceleration from the longitudinal acceleration signal between the time point determined to be the heel contact operation and the time point determined to be the sole contact operation. It can be obtained by calculation. The same applies to longitudinal acceleration or vertical acceleration in other periods.

  The average acceleration calculated in this way can be adopted as an estimation index, and the lower limb strength can be measured from the acceleration-lower limb strength relationship. Specifically, the CPU 24 calculates the average acceleration calculated after the acceleration-lower limb strength relationship as shown in FIGS. 8 to 13 stored in the ROM 26 (after longitudinal acceleration in a specific period corresponding to the acceleration-lower limb strength relationship). Alternatively, the lower limb strength can be measured by deriving the lower limb strength corresponding to the average acceleration. The lower limb muscle strength derived at this time can be temporarily stored in the RAM 28, and can be stored in the recording unit 50 automatically or by a user operation. Further, the derived lower limb strength or the past measurement result recorded in the recording unit 50 is displayed on the display unit 40 together with information such as user information and date stored in the recording unit 50, for example. May be.

  Further, in the acceleration-lower limb muscle strength relationship shown in FIGS. 8 to 13 or the acceleration-lower limb muscle strength relationship obtained from the results of the multiple regression analysis described above, two or more dorsiflexion forces and knee extension forces can be derived respectively. It is also possible to derive the dorsiflexion force and the knee extension force by selecting the acceleration-lower limb muscle strength relationship. The derived dorsiflexion force and knee extension force can be temporarily stored in the RAM 28 and stored in the recording unit 50 automatically or by a user operation. In addition, when storing in the recording unit 50, information such as a user and date, and information on the relationship between acceleration and lower limb strength used respectively may be stored.

  Further, the current walking age of the user can also be calculated from the strength of the derived lower limb muscle strength. Specifically, the ROM 26 stores in advance the relationship between the strength of the lower limb muscle strength and the walking age, and calculates the walking age by applying the calculated lower limb strength to the relationship between the lower limb muscle strength and the walking age. can do. In addition, as the lower limb muscle strength used when calculating the walking age, at least one of dorsiflexion force and knee extension force can be employed. The calculated walking age is temporarily stored in the RAM 28. Furthermore, the walking age determination result may be stored in the recording unit 50 automatically or by user operation.

  As described above, when the strength of the lower limb muscle strength is derived by the CPU 24, the result can be displayed on the display unit 40. In this case, the user can select an item to be displayed from the dorsiflexion force, the knee extension force, the walking age, and the like, and the determination result of the selected item can be displayed. Alternatively, all the determination results may be automatically displayed. Further, by displaying an exercise method effective for improving the lower limb muscle strength on the display unit based on the estimated lower limb muscle strength and walking age, it is possible to teach the user the exercise guidelines.

  As described above, according to the walking analysis device 100 of the present embodiment, the user's lower limb muscle strength can be measured without using a large-scale device. Further, since the user only has to walk with the walking analysis device 100 of the present embodiment attached to the waist, the user is not burdened excessively. Furthermore, in the gait analysis apparatus 100 of the present embodiment, two kinds of lower limb muscle strengths of dorsiflexion force and knee extension force can be measured only by walking motion, and therefore separate measurements are performed to measure these lower limb muscle strengths. There is no need to do the way. Furthermore, when the gait analysis device 100 of the present embodiment is used to improve the deterioration of the lower limbs, information on more specific lower limb muscle strength such as dorsiflexion force and knee extension force can be obtained. The person can know in detail which part of the lower limb is declining. Therefore, it can be determined relatively easily what kind of means should be taken to improve the deterioration of the lower limbs.

  (3rd Embodiment) Next, 3rd Embodiment which estimates walking speed as walking ability is described. As a result of intensive studies by the present inventors, there is a correlation between a specific estimated index calculated based on a temporal change in the acceleration of the waist and a walking speed in a specific period in which a specific walking motion is performed during walking. It was found that there was a relationship.

  That is, the walking analysis apparatus 100 according to the third embodiment includes acceleration measurement means (accelerometer 10) that detects accelerations in the vertical direction, the front-rear direction, and the horizontal direction at the waist during walking, and at least one of the accelerations. A period extracting means (ROM 26, CPU 24) for extracting a specific period in which a specific walking motion related to the walking speed is performed based on a time change of two accelerations, and a time of at least one of the accelerations in the specific period And an estimated index calculating means (ROM 26, CPU 24) for calculating an estimated index related to the walking speed based on the change. Furthermore, it has walking ability estimation means (ROM 26, CPU 24) for estimating the walking speed using the estimated index calculated by the estimated index calculating means and the relationship between the estimated index prepared in advance and the walking speed. Here, the walking ability estimation means may include storage means (storage device, RAM 28, ROM 26) in which the relationship between the estimated index prepared in advance and the walking speed is stored.

  The relationship (relational expression) between the estimated index stored in the ROM 26 and the walking speed will be described below. First, each estimated index and the relationship (relational expression) between each estimated index and walking speed will be described in detail.

  (Estimated index V1) Although the forward speed, that is, the walking speed can be calculated to some extent by integrating the longitudinal acceleration, the influence of the integration error is large and there is a problem in the estimation accuracy. Therefore, I thought as follows. Walking motion is a repetition of acceleration and deceleration operations. When you look at the ratio of the time to accelerate (acceleration time) and the time to decelerate (deceleration time), the longer the time to accelerate, the faster the walking We thought that the ratio of deceleration time in one step time can be used as an estimation index because it tends to be speed.

  FIG. 14 is an explanatory diagram illustrating one step time and deceleration time using a graph similar to the graph of FIG. In this case, the deceleration time corresponds to the time from the point at which the maximum speed is obtained in the front-rear direction (referred to as the front-rear maximum speed position) to the point at which the minimum speed in the front-rear direction (referred to as the front-rear minimum speed position). Here, the following formula was used as the estimated index V1.

  Estimated index V1 = Time from the maximum front / rear speed position to the lowest front / rear speed position / step time.

  Acceleration is a derivative of velocity. Therefore, in FIG. 14, the point where the front and rear speeds are maximum is where the acceleration crosses the time axis from positive to negative, and the point where the front and rear speeds are minimum is where the acceleration crosses the time axis from negative to positive. Therefore, “the time from the maximum speed position to the minimum speed position” is “the time when the longitudinal acceleration becomes negative to positive“ − ”the time when positive and negative becomes positive”.

  Using the explanatory diagram of FIG. 14, the deceleration time and the like were calculated, and the graph of FIG. 15 was created. The graph of FIG. 15 is a graph in which the estimated index V1 (deceleration time / step time) (%) is taken on the X axis, and the walking speed / height (m / sec / m) is taken on the Y axis. The actual walking speed was used as the walking speed on the Y axis. The actual walking speed was obtained using a 3D motion analysis system. In order to eliminate individual differences and normalize, the walking speed divided by height on the Y axis was used. As shown in FIG. 15, it can be seen that the smaller the rate of deceleration time in one step time, the larger the value obtained by dividing the walking speed by the height. As shown in FIG. 15, a negative correlation was found between the index V1 and the value obtained by dividing the actually measured cheek speed by the height.

  (Estimated index V2) When a large positive acceleration occurs, a large negative acceleration occurs. When the walking speed is fast, the acceleration change is considered large. Therefore, the negative longitudinal acceleration indicating deceleration during the one-step period was considered to reflect the walking speed.

  When the acceleration change is large, the energy consumption increases. Since the estimated index V1 and the walking speed are correlated as described above, it can be seen that when the walking speed is high, the time spent for the acceleration operation becomes longer in one step time. In other words, when large braking (negative acceleration) occurs, it is considered that the acceleration time was long. This is considered to be walking such that the energy consumption does not increase as much as possible, and it is considered that the normal walking is based on the assumption that the movement is performed with the minimum energy consumption. The estimated index V2 is shown below.

  Estimated index V2 = Integral value of the longitudinal acceleration / integral period time from the point at which the longitudinal acceleration changes from positive to negative to the point at which negative to positive changes.

  "The integrated value of the longitudinal acceleration from the point at which the longitudinal acceleration changes from positive to negative to the point at which it changes from negative to positive" means the longitudinal acceleration in the period representing the "time from the maximum speed position to the minimum speed position" of the estimated index V1 Is a value obtained by integrating over time. That is, the estimated index V2 is a value obtained by dividing the integral value by the integration period time, and means an average deceleration during the deceleration operation. FIG. 16 is an explanatory diagram illustrating the integrated value of the longitudinal acceleration from the point at which the longitudinal acceleration changes from positive to negative to the point at which it changes from negative to positive, using a graph similar to the graph of FIG.

  FIG. 17 shows a graph in which the estimated index V2 (m / sec / sec) is taken on the X axis and walking speed / height (m / sec / m) is taken on the Y axis. As shown in FIG. 17, there was a negative correlation between the estimated index V2 and the walking speed / height.

  (Estimated index V3) The body center of gravity during walking repeatedly moves up and down. The position of the center of gravity of the body is lowest when the heel is touched, and is highest when the body is in an upright state (mid stance). After the middle stage of stance, when the foot is swung forward, the center of gravity of the body moves downward and becomes the lowest point when the heel touches.

  Therefore, the upward acceleration generated when the center of gravity moves upward, the upward deceleration acceleration generated when the upward moving speed is decelerated, and the downward acceleration generated when the center of gravity moves downward also reflect the walking speed. it is conceivable that. Since the upward acceleration is considered to be caused by vibration due to the impact at the time of ground contact, attention was paid to the upward deceleration acceleration and downward acceleration, which are considered to be less affected by the vibration due to external force.

  The middle stance phase appears after the toes leave. Therefore, we thought that the negative acceleration in the vertical acceleration after the toe-off shows the vertical deceleration acceleration just before the middle stance and the downward acceleration of the body center of gravity by swinging the foot forward. The positive vertical acceleration immediately before the heel contact was considered to be an acceleration for decelerating the downward speed to mitigate the impact force at the heel contact. This operation of impact relaxation can be said to be based on the assumption that muscle load at the time of saddle contact (braking period) can be reduced and energy consumption is suppressed. The sum of absolute values of both integral values considered to reflect the walking speed was used as the estimated index V3.

  Estimated index V3 = | Deceleration amount of upper speed / period time | + | Deceleration amount of lower speed / period speed |

  The “deceleration amount of the upward speed” indicates a time integral value of the vertical acceleration over a period from the tip of the foot to the point where the vertical acceleration changes from negative to positive. Therefore, | the amount of deceleration of the upward speed / period time | means the absolute value of the average deceleration in the vertical direction in this specific period. Further, the “deceleration amount of the downward speed” indicates a time integration value of the vertical acceleration with a specific period from the point at which the vertical acceleration changes from negative to positive to the heel contact. Therefore, | the amount of deceleration of the downward speed / period time | means the absolute value of the average acceleration in the vertical direction during this specific period.

  FIG. 18 is an explanatory diagram illustrating a braking period, an acceleration period, and a late stance phase using a graph similar to the graph of FIG. The deceleration amount of the upward speed and the deceleration amount portion of the downward speed are represented by arrows. FIG. 19 shows a graph in which the estimated index V3 (m / sec / sec) is taken on the X axis and the walking speed / height (m / sec / m) is taken on the Y axis. As shown in FIG. 19, there was a positive correlation between the estimated index V3 and the walking speed / height.

  (Estimated index V4) Estimated index V4 = (Integrate the longitudinal acceleration from the right [left] 踵 ground contact to the left [right] 踵 ground, the difference between the maximum speed and the minimum speed) / (the minimum before and after the maximum speed position before and after Time to speed position / step time).

  FIG. 20 is a diagram illustrating a point at which the maximum speed and a minimum speed at one step time are reached. FIG. 21 shows a graph in which the estimated index V4 (m / sec / sec) is taken on the X axis and the walking speed / height (m / sec / m) is taken on the Y axis. As shown in FIG. 21, the estimated index V4 and the walking speed / height showed a positive correlation.

  (Estimated index V5) Next, the longitudinal acceleration from the right foot apex, which is the first acceleration phase, to the late right stance phase (the point where the vertical acceleration after the foot apex is changed from negative to positive) is integrated. The speed was set as the walking speed estimation index V5. (Similarly, the longitudinal velocity in the late stance phase of the left stance can also be used) Estimated index V5 = integral value of the longitudinal acceleration from the right [left] toe separation to the right [left] stance phase.

  FIG. 22 is an explanatory diagram illustrating a braking period, an acceleration period, and a late stance phase using a graph similar to the graph of FIG. The shaded area is the value of the estimated index V5. The estimated index V5 was calculated using the explanatory diagram of FIG. 22, and the graph of FIG. 23 was created. FIG. 23 shows a graph with the estimated index V5 (m / sec) on the X axis and the walking speed / height (m / sec / m) on the Y axis. As shown in FIG. 23, there was a positive correlation between the estimated index V5 and walking speed / height.

  (Estimated index V6) Next, the velocity calculated by integrating the vertical acceleration from the early stage of stance (the first point at which the vertical acceleration changes from positive to negative after the toe off) to the late stage of stance,推定 The sum of the speeds calculated by integrating the vertical acceleration until contact with the ground was used as the estimated index V6.

  Estimated index V6 = | the sum of the integrated value of the vertical acceleration from the initial stance phase of one foot to the late stance phase and the integrated value of the vertical acceleration from the late stance phase of one foot to immediately before the heel contact of the other foot.

  FIG. 24 is an explanatory diagram illustrating the braking period, the acceleration period, the initial stance phase, and the late stance phase using a graph similar to the graph of FIG. The estimated index V6 was calculated using the explanatory diagram of FIG. 24, and the graph of FIG. 25 was created. FIG. 25 shows a graph with the estimated index V6 (m / sec) on the X axis and the walking speed / height (m / sec / m) on the Y axis. As shown in FIG. 25, there was a positive correlation between the estimated index V6 and walking speed / height.

  (Estimated index V7) Estimated index V7 = estimated index V6 / estimated index V1.

  The estimated index V7 was calculated, and the graph of FIG. 26 was created. FIG. 26 shows a graph in which the estimated index V7 (m / sec / sec) is taken on the X axis and the walking speed / height (m / sec / m) is taken on the Y axis. As shown in FIG. 26, there was a positive correlation between the estimated index V7 and walking speed / height.

  (Walking speed estimation formula = Relational expression between estimated index V and walking speed) The relational expression for estimating the walking speed was determined using the estimated indexes V1 to V7. The estimated speeds V1 to V7 can determine the walking speed even if using coefficients independently. In this case, a relational expression represented by a straight line shown in FIGS. 15, 17, 19, 21, 23, 25, and 26 can be used. Further, in order to improve accuracy, a relational expression can be obtained by performing a multiple regression analysis using a plurality of estimation indices. Examples of estimation equations using the estimation indices V1 to V3 are shown below.

  Walking speed = 0.249 × estimated index V1-0.091 × estimated index V2 + 0.049 × estimated index V3 + 0.269.

  The estimated walking speed was calculated by substituting the value calculated from each measurement data into the above relational expression, and the correlation with the actually measured walking speed was observed. FIG. 27 shows a graph in which the estimated walking speed / height (m / sec / m) is taken on the X axis and the actually measured walking speed / height (m / sec / m) is taken on the Y axis. As shown in FIG. 27, a high positive correlation was observed. The actually measured walking speed is the same as the walking speed used in FIGS. Further, for example, a relational expression using estimated indices V4 to 7 is shown below.

  Walking speed = 0.18 × estimated index V4 + 0.56 × estimated index V5 + 0.69 × estimated index V6-0.15 × estimated index V7 + 0.38.

  The estimated walking speed was calculated by substituting the value calculated from each measurement data into the above relational expression, and the correlation with the actually measured walking speed was observed. FIG. 28 shows a graph in which the estimated walking speed / height (m / sec / m) is taken on the X axis and the actually measured walking speed / height (m / sec / m) is taken on the Y axis. As shown in FIG. 28, a high positive correlation was observed. The actually measured walking speed is the same as the walking speed used in FIGS.

  Hereinafter, the operation and operation of the walking analysis apparatus 100 according to the third embodiment will be described. First, when the walking analysis apparatus 100 of the present embodiment is mounted on the waist and walking is started, the accelerometer 10 of the walking analysis apparatus 100 detects the longitudinal acceleration, the lateral acceleration, and the vertical acceleration of the waist. The accelerometer 10 outputs the detected time change of acceleration as a waveform of an electric signal (acceleration signal). The acceleration signal output from the accelerometer 10 is digitized by the A / D converter 22 and temporarily stored in the RAM 28.

  When an acceleration signal (longitudinal acceleration, lateral acceleration, vertical acceleration) corresponding to one cycle of walking motion (may be several cycles) is stored in the RAM 28, the walking motion determination means configured by the CPU 24 and the ROM 26 The timing and period (specific period) at which a specific walking motion is performed are extracted from the time change of the acceleration signal. Specifically, it is possible to detect the peak of the acceleration signal by differentiating the acceleration signal with respect to time and determine that a specific walking motion starts or ends at the peak. Alternatively, focusing on one acceleration selected from longitudinal acceleration, vertical acceleration, and lateral acceleration, it is possible to determine that a specific walking motion starts or ends when the acceleration changes from positive to negative or from negative to positive. it can. Alternatively, it is possible to determine the timing when the acceleration becomes a predetermined ratio with respect to the acceleration at the peak of the acceleration signal as the start time and the end time of the walking motion. Furthermore, among all the period extraction methods described here, the most appropriate one may be selected at any time, or a method in which these period extraction methods are appropriately combined may be employed.

  As described above, if the time point when the specific walking motion is performed can be grasped, the CPU 24 calculates an estimated index related to the walking speed during the specific period in which the specific walking motion is performed. Specifically, the CPU 24 can measure the walking speed by applying each calculated estimated index data to the estimated index-walking speed relationship stored in the ROM 26. The walking speed derived at this time can be temporarily stored in the RAM 28, and can be stored in the recording unit 50 automatically or by a user operation. Further, the derived walking speed or the past measurement result recorded in the recording unit 50 is displayed on the display unit 40 together with information such as user information and date stored in the recording unit 50, for example. May be.

  (4th Embodiment) Next, embodiment in the case of estimating a stride as walking ability is described. As a result of intensive studies by the present inventors, there is a correlation between the specific estimated index calculated based on the temporal change of the waist acceleration and the stride in a specific period in which a specific walking motion is performed during walking. It was found that there is.

  That is, the walking analysis apparatus 100 according to the fourth embodiment includes acceleration measuring means (accelerometer 10) that detects accelerations in the vertical direction, the front-rear direction, and the horizontal direction at the waist during walking, and at least one of the accelerations. Period extracting means (ROM 26, CPU 24) for extracting a specific period during which a specific walking motion related to the stride is performed based on a time change of two accelerations, and a time change of at least one of the accelerations in the specific period And an estimated index calculating means (ROM 26, CPU 24) for calculating an estimated index related to the stride. Furthermore, it has walking ability estimation means (ROM 26, CPU 24) for estimating the stride using the estimated index calculated by the estimated index calculation means and the relationship between the estimated index prepared in advance and the stride. Here, the walking ability estimation means may include storage means (storage device, RAM 28, ROM 26) in which the relationship between the estimated index and the stride prepared in advance is stored.

  A relationship (relational expression) between the estimated index and the stride stored in the ROM 26 will be described below. In estimating the stride, it is assumed that the normal walking is an exercise performed with the minimum energy as in the third embodiment.

  (Estimated index S1) Consider steady walking with no speed change from the time of heel contact to the time of the next heel contact. There is a positive correlation between the walking speed and the stride, and the stride is large when the walking speed is fast, and the stride is small when the walking speed is slow. If the walking speed is fast, a large acceleration occurs, and a large deceleration acceleration is also generated. Therefore, it can be said that the deceleration acceleration amount in one step period reflects the stride. In addition, walking includes walking with a small crotch that has a high walking speed and a large crotch that has a slow walking speed. The amount of deceleration was considered to be the amount of deceleration that better reflected the stride. In other words, to accelerate the decelerated speed to the initial speed, the body is moved forward by the action of the triceps surae. Of course, in order for the body to move forward, it is necessary to swing the foot forward. I thought that if the amount of deceleration was large, it would be necessary to accelerate forward and swing out a large foot accordingly. Therefore, the longitudinal acceleration from the sole contact to the toe-off is integrated, and a value obtained by dividing the integrated value by the period time is defined as the estimated index S1.

  Estimated index S1 = time integrated value / period time of longitudinal acceleration from right [left] sole contact to left [right] foot tip separation.

  The time integral value / periodic time of the longitudinal acceleration from the right [or left] sole contact to the left [or right] foot-tip separation represents the average deceleration during that period.

  FIG. 29 is an explanatory diagram illustrating the contact between the ground contact and the tip of the toes using a graph similar to the graph of FIG. The estimated index S1 was calculated using the explanatory diagram of FIG. 29, and the graph of FIG. 30 was created. FIG. 30 shows a graph with the estimated index S1 (m / sec / sec) on the X axis and the stride / height (m / m) on the Y axis. The stride in this case was measured using a three-dimensional motion analysis system and a floor reaction force meter. As shown in FIG. 30, there was a negative correlation between the estimated index S1 and the stride / height.

  (Estimated index S2) The body center of gravity during walking repeats vertical movement. The position of the center of gravity of the body is lowest when the heel is touched, and is highest when the body is in an upright state (mid stance). After the middle stage of stance, when the foot is swung forward, the center of gravity of the body moves downward and becomes the lowest point when the heel touches. Since there is a positive correlation between walking speed and stride, the longer the stride, the shorter the time from heel contact to mid-stance, and the upward acceleration that occurs when the center of gravity moves upward, and the deceleration of the upward movement speed The upward deceleration acceleration generated at the time is also large. In addition, if the amount of swinging forward is large, the downward acceleration that occurs when the center of gravity moves downward also increases. Therefore, it is considered that the upward acceleration, the upward deceleration acceleration, and the downward acceleration reflect the stride. Here, the upward acceleration is considered to be caused by the vibration caused by the impact at the time of contact with the heel. Therefore, attention is paid to the upward deceleration acceleration and the downward acceleration which are considered to be less affected by the vibration due to the external force.

  The middle stance phase appears after the toes leave. Therefore, the negative acceleration seen in the displacement of the vertical acceleration after the toe-off was considered to indicate the vertical deceleration acceleration just before the middle stance and the downward acceleration of the body center of gravity by swinging the foot forward. The positive vertical acceleration immediately before the heel contact was considered to be an acceleration for decelerating the downward speed to mitigate the impact force at the heel contact. The sum of the absolute values of both integral values considered to reflect the stride was used as the estimated index S2. The estimated index S2 is the same as the estimated index V3.

  Estimated index S2 = | Deceleration amount of upper speed / period time | + | Deceleration amount of lower speed / period speed |

  The “deceleration amount of the upward speed” indicates a time integral value of the vertical acceleration over a period from the tip of the foot to the point where the vertical acceleration changes from negative to positive. Therefore, | the deceleration amount of the upward speed / period time | means the absolute value of the vertical average deceleration in this period. Further, the “deceleration amount of the downward speed” indicates a time integral value of the vertical acceleration over a period from the point at which the vertical acceleration changes from negative to positive to the heel contact. Therefore, | the amount of deceleration of the downward speed / period time | means the absolute value of the vertical average acceleration in this period.

  FIG. 18 is an explanatory diagram illustrating a braking period, an acceleration period, and a late stance phase using a graph similar to the graph of FIG. The deceleration amount of the upward speed and the deceleration amount portion of the downward speed are represented by arrows. FIG. 31 shows a graph with the estimated index S2 (m / sec / sec) on the X axis and the stride / height (m / m) on the Y axis. As shown in FIG. 31, there was a positive correlation between the estimated index S2 and the stride / height.

  (Estimated index S3) For the operation in the one-step period (period from heel contact to the other heel contact), deceleration during the period (both leg support period) from the heel contact to the other toe-off (the toes away from the ground) There are movements and forward acceleration movements in the single leg support period after the toes take off. The forward acceleration action swings out the foot on the free leg side forward and creates forward acceleration by the action of the triceps surae. When the stride is small, the amount of swinging forward is small. Also, during the single leg support period, it is necessary to maintain the stability of the body with one leg, and the burden on the lower limb muscles is large. For this reason, when the amount of swinging the foot forward is small, the swinging time is considered to be short. Based on the above assumption, it is difficult to think of creating a large acceleration amount in a short time. Therefore, when the stride is small, that is, when the free leg period is short, it is conceivable that the acceleration amount produced in the single leg support period decreases. To compensate for this decrease in acceleration, we thought that the amount of deceleration during the two-leg support period could be reduced. As a result, the speed undulation during walking is considered to be small, and the longitudinal acceleration from the heel-contact to the other heel-contact is integrated and converted into a speed, and the difference between the maximum speed and the minimum speed is set as the estimated index S3.

  Estimated index S3 = difference between the maximum speed and the minimum speed from right [left] 踵 grounding to left [right] grounding.

  "The difference between the maximum speed and the minimum speed from right [left] 踵 grounding to left [right] 踵 grounding" means the longitudinal acceleration from "right [left] 踵 grounding to left [right] 踵 grounding in the estimated index V4. Integrate and indicate the same as the difference between the maximum speed and the minimum speed ". From the value obtained by time-integrating the longitudinal acceleration (maximum speed) in the period from the ground contact to the point where the longitudinal acceleration becomes positive to negative, It is a value obtained by subtracting a value (minimum speed) obtained by time-integrating the longitudinal acceleration in a period from the negative to the positive.

  FIG. 32 shows a graph with the estimated index S3 (m / sec / sec) on the X axis and the stride / height (m / m) on the Y axis. As shown in FIG. 32, there was a positive correlation between the estimated index S3 and the stride / height.

  (Estimated index S4) When the stride is small, the amount of swinging the foot forward is small. During the period in which the foot is swung forward (the swing leg period), it is necessary to maintain the stability of the body with a single leg, and it is considered that the burden on the lower limb muscles is also great. Based on the above assumption, if the amount of swinging out the foot is small, it is considered that the period of the swing phase is also short. Therefore, the time ratio of the swing leg period in one step time is considered to reflect the stride, and the stride estimation index S4 is used.

  Estimated index S4 = time ratio from heel-contact to foot-point separation with respect to one step time. FIG. 33 is an explanatory diagram illustrating heel-contact and foot-toe separation using a graph similar to the graph of FIG.

  (Strength Estimating Formula = Relational Formula Between Stride and Estimated Index S) A person with a fast walking speed tends to have a wide stride. Therefore, the relational expression for estimating the stride is determined using the estimated indices S1 to S4 or the estimated indices V1 to V7. Even if each estimated index S1-S4 and estimated index V1-V7 are used independently, walking speed can be calculated | required. In this case, a relational expression represented by a straight line shown in FIGS. In order to improve accuracy, multiple regression analysis was performed using multiple estimation indices, and a relational expression with the following coefficients was found. For example, a relational expression using estimated indices S1 to S4 is shown below.

  Step length = 0.163 × estimated index S1-0.023 × estimated index S2 + 0.286 × estimated index S3 + 0.027 × estimated index S4-0.236

  The estimated stride was calculated by substituting the value calculated from each measurement data into the above relational expression, and the correlation with the actually measured stride was observed. FIG. 34 shows a graph in which the estimated stride / height (m / m) is taken on the X axis and the actually measured stride / height (m / m) is taken on the Y axis. As shown in FIG. 34, a high positive correlation was observed. Further, for example, relational expressions using estimated indexes V4, V6, and V7 are shown below.

  Step length = 0.11 × estimated index V4 + 0.27 × estimated index V6-0.06 × estimated index V7 + 0.24.

  The estimated stride was calculated by substituting the value calculated from each measurement data into the above relational expression, and the correlation with the actually measured stride was observed. FIG. 35 is a graph in which the estimated stride / height (m / m) is taken on the X axis and the actually measured stride / height (m / m) is taken on the Y axis. As shown in FIG. 24, a high positive correlation was observed.

  As described above, according to the embodiment of the present invention, since only the acceleration of the waist is measured, it is possible to easily measure without requiring specialized knowledge. There is no particular burden on the subject. Also, by extracting a specific period during which a specific walking motion is performed and calculating an estimated index related to the walking speed or stride based on the temporal change in acceleration during that specific period, a more universal walking speed or stride Estimation can be performed.

  Hereinafter, the operation and operation of the walking analysis apparatus 100 according to the fourth embodiment will be described. First, when the walking analysis apparatus 100 of the present embodiment is mounted on the waist and walking is started, the accelerometer 10 of the walking analysis apparatus 100 detects the longitudinal acceleration, the lateral acceleration, and the vertical acceleration of the waist. The accelerometer 10 outputs the detected time change of acceleration as a waveform of an electric signal (acceleration signal). The acceleration signal output from the accelerometer 10 is digitized by the A / D converter 22 and temporarily stored in the RAM 28.

  When an acceleration signal (longitudinal acceleration, lateral acceleration, vertical acceleration) corresponding to one cycle of walking motion (may be several cycles) is stored in the RAM 28, the walking motion determination means configured by the CPU 24 and the ROM 26 The timing and period (specific period) at which a specific walking motion is performed are extracted from the time change of the acceleration signal. Specifically, it is possible to detect the peak of the acceleration signal by differentiating the acceleration signal with respect to time and determine that a specific walking motion starts or ends at the peak. Alternatively, focusing on one acceleration selected from longitudinal acceleration, vertical acceleration, and lateral acceleration, it is possible to determine that a specific walking motion starts or ends when the acceleration changes from positive to negative or from negative to positive. it can. Alternatively, it is possible to determine the timing when the acceleration becomes a predetermined ratio with respect to the acceleration at the peak of the acceleration signal as the start time and the end time of the walking motion. Furthermore, among all the period extraction methods described here, the most appropriate one may be selected at any time, or a method in which these period extraction methods are appropriately combined may be employed.

  As described above, if the time point when the specific walking motion is performed can be grasped, the CPU 24 calculates an estimated index related to the stride during the specific period in which the specific walking motion is performed. Specifically, the CPU 24 can measure the stride by applying each estimated index data calculated to the estimated index-stride relationship stored in the ROM 26. The stride derived at this time can be temporarily stored in the RAM 28, and can be stored in the recording unit 50 automatically or by a user operation. Further, the derived stride or the past measurement result recorded in the recording unit 50 is displayed on the display unit 40 together with information such as user information and date stored in the recording unit 50, for example. Also good.

  Furthermore, the user's risk of falling can be determined from the walking speed and the stride derived in the third and fourth embodiments. The pace (step / min) is calculated from the derived walking speed (m / sec) and step length (m). The gait is determined by the gait (step / min) = walking speed (m / sec) / step length (m) × 60. The fall risk of the user can be determined by a combination of the obtained walking speed, step length, and pace value.

  For example, as illustrated in FIGS. 36 and 37, the walking speed is 0.69 (m / sec) or less, 1.5 (m / sec) or more, and the interval is divided into multiple stages. Also, the stride is 0.49 (m) or less, 0.8 (m) or more, and the interval is divided into a number of stages. The pace is divided into two steps of 90 (step / min) or less and 91 (step / min) or more. The fall risk can be divided into, for example, four stages (fall danger, fall attention, whip attention, no problem) by combining the walking speed, the stride, and the pace. For example, if the walking speed is 0.69 (m / sec) or less, the step length is 0.49 (m) or less, and the pace is 90 (step / min) or less, it is determined that there is a risk of falling. The numerical values given as examples can be appropriately set according to the user's physique, age, gender, walking state (normal walking, full power walking, etc.) and the like. At this time, normal walking means walking as usual, and full power walking means walking as fast as possible.

  More specifically, the ROM 26 stores in advance the relationship between the fall risk and the walking speed, the stride and the pace, and the calculated walk speed, the stride and the step are calculated as the fall risk, the walking speed, the stride and the pace, respectively. By applying this relationship, the risk of falls can be determined. The determined fall risk is temporarily stored in the RAM 28. Furthermore, the fall risk determination result may be stored in the recording unit 50 automatically or by user operation.

  As described above, when the walking speed, the stride, the pace, and the fall risk are derived by the CPU 24, the results can be displayed on the display unit 40. In this case, the user can select an item to be displayed from the walking speed, the stride, the pace, the fall risk, and the like, and the result of the selected item can be displayed. Alternatively, all the results may be automatically displayed. Furthermore, by displaying an exercise method effective for improvement on the display unit based on the determined fall risk, it is possible to teach the user the exercise guidelines.

  For example, an effective exercise menu can be displayed as an exercise guideline. As an example of the exercise menu, an example of a one-leg chair standing exercise for strengthening the lower limb muscle strength is shown. This displays an explanatory diagram of the exercise, and displays the explanation in an easy-to-understand manner even in text. As an explanation, “Stand up with one foot from a sitting chair without holding a handrail and with enough weight on the soles of your feet. When you get used to it, try to use a lower chair. If there is, let's get up with your hands on your knees. "

  Furthermore, in the embodiment of the present invention (first, second, third, and fourth embodiments), when extracting the specific period during walking, the specific period in each of the left and right feet is extracted, In each of the above, an estimated index is calculated, and the walking ability of each of the left and right feet can be estimated based on the calculated estimated index. In this case, in order to determine which of the left and right feet is in the specific period during which the walking motion is performed, by detecting whether the time change of the acceleration is positive or negative in the time change of the left and right acceleration. Judgment can be made. It can be set as appropriate whether the positive or negative acceleration corresponds to the left or right foot.

  In this way, the dorsiflexion force, lower limb muscle strength, walking speed and stride of each leg can be estimated, and when used for rehabilitation, etc., the rehabilitation effect, rehabilitation necessary parts, etc. are clarified and efficient treatment. Can be done.

  Furthermore, the present invention is not limited to the above-described embodiment, but includes the estimation of other walking ability. For example, walking ability includes steps (distance of left and right feet in the left and right direction), movable range angle of each indirect (hip joint, knee joint, ankle joint), torque of each joint, extension / flexion / extension strength of each joint, floor Reaction force etc. can be illustrated. Further, the left / right balance of each walking ability and the distortion of the lower limb skeleton may be numerically estimated as walking ability separately.

  Further, in the present invention, the result of estimating the walking ability by the method as described above is taught to the user, and a guideline (exercise guideline or the like) for improving the walking ability may be taught to the user. . For example, the level of walking ability is classified in stages, exercise guidelines corresponding to the level are prepared in advance, the level is divided from the estimated walking ability, and the exercise guide corresponding to the level is displayed on the display unit 40. May be displayed. Other means may be employed as means for teaching a guideline for improving walking ability.

  It is also possible to construct a system that utilizes the results of the above-described walking analysis device. For example, configuring a health maintenance / promotion system having information storage means for storing walking ability information determined based on three-dimensional acceleration and point calculation means for calculating points based on the walking ability information. I can do it.

Next, a health maintenance / promotion system using the analysis result of the walking analysis device 100 will be described. The walking analysis device 100 can determine the risk of falling, which is one of the walking ability of the subject, based on the three-dimensional acceleration. The fall risk is classified as, for example, the following five stages and stored as information.
(1) Falling risk (2) Falling attention (3) Whispering attention (4) No problem at this stage (step length reduction)
(5) No problem at all

This fall risk is converted into points in the CPU 24 as follows.
(1) Falling risk: -2 points (2) Falling precautions: -1 point (3) Whispering attention: 0 points (4) No problem at this stage (step length reduction): +1 point (5) No problem at all: +2 points

  As shown in FIG. 38, the gait analyzer 100 is connected to a personal computer 200 via a well-known USB. Then, an input screen as shown in FIG. 39 is displayed on the personal computer 200 so that the name, insurer number, and telephone number of the subject can be input. When the OK button is clicked after all items are input, the gait analyzer 100 sends the input information and the fall risk points to the health insurance association host computer 400 via the Internet 300. The host computer 400 updates the management ledger already created using the fall risk points of the subject transferred this time.

  For example, if the information transmitted through the net is a fall risk point of “Kikuko Hayashi” (current: +1 point), the fall point shown in FIG. 40 is “+ 2 ”. When the cumulative fall risk points reach +10 points, the insurance union will give the subjects more motivation to maintain health and promote health if it is set to give prizes to Kikuko Hayashi. it can.

  In this case, as described above, the fall risk as the walking ability is determined based on the three-dimensional acceleration acting on the walking analysis device 100. Therefore, even if the walking analysis device 100 is held and shaken, the three-dimensional acceleration is not detected. Since it does not work, data tampering is prevented.

  In other words, health maintenance enhancement having information storage means for storing walking ability information determined based on three-dimensional acceleration for each user and point calculation means for calculating points based on the walking ability information By configuring the system, the reliability of point grant can be improved.

The block diagram which shows an example of the walk analysis apparatus of this invention. The figure which shows an example of an acceleration-dorsiflexion force relationship. The figure explaining walking operation. The figure explaining the method of prescription | regulation of the strength of a dorsiflexion force in 1st Embodiment. The figure which shows the time change waveform of an acceleration signal in 1st Embodiment. In 1st Embodiment, the figure explaining the difference in the dorsiflexion angle at the time of the walk in case dorsiflexion force differs. The figure which shows the time change waveform of the acceleration signal in embodiment, and corresponding walking motion. The figure which shows an example of acceleration-lower limb muscle strength relationship in 2nd Embodiment. The figure which shows an example of acceleration-lower limb muscle strength relationship in 2nd Embodiment. The figure which shows an example of acceleration-lower limb muscle strength relationship in 2nd Embodiment. The figure which shows an example of acceleration-lower limb muscle strength relationship in 2nd Embodiment. The figure which shows an example of acceleration-lower limb muscle strength relationship in 2nd Embodiment. The figure which shows an example of acceleration-lower limb muscle strength relationship in 2nd Embodiment. Explanatory drawing which illustrated one step time and deceleration time in 3rd Embodiment. The figure which shows an example of the relationship between the estimation parameter | index V1 and walking speed / height in 3rd Embodiment. Explanatory drawing which illustrated the integrated value of the longitudinal acceleration from the point from which a longitudinal acceleration changes from positive to negative to the point from negative to positive in 3rd Embodiment. The figure which shows an example of the relationship between the estimation parameter | index V2 and walking speed / height in 3rd Embodiment. In 3rd Embodiment, explanatory drawing which illustrated the braking period, the acceleration period, and the late stage of stance. The figure which shows an example of the relationship between the estimation parameter | index V3 and walking speed / height in 3rd Embodiment. The figure explaining the point used as the maximum speed and the minimum speed in one step time in 3rd Embodiment. The figure which shows an example of the relationship between the estimation parameter | index V4 and walking speed / height in 3rd Embodiment. Explanatory drawing which illustrated the range which the braking period, the acceleration period, the stance phase, and the estimation parameter | index V5 show in 3rd Embodiment. The figure which shows an example of the relationship between the estimation parameter | index V5 and walking speed / height in 3rd Embodiment. Explanatory drawing which illustrated the range which the braking period, the acceleration period, the initial stage of stance, the late stage of stance, and the estimation parameter | index V6 show in 3rd Embodiment. The figure which shows an example of the relationship between the estimation parameter | index V6 and walking speed / height in 3rd Embodiment. The figure which shows an example of the relationship between the estimation parameter | index V7 and walking speed / height in 3rd Embodiment. The figure which shows an example of the relationship between estimated walking speed / height and measured walking speed / height. The figure which shows an example of the relationship between estimated walking speed / height and measured walking speed / height. Explanatory drawing which illustrated plantar grounding and toe-off. The figure which shows an example of the relationship between estimation parameter | index S1 and stride / height. The figure which shows an example of the relationship between estimation parameter | index S2 and stride / height. The figure which shows an example of the relationship between estimation parameter | index S3 and stride / height. Explanatory drawing which illustrated heel grounding and toe-off. The figure which shows an example of the relationship between presumed stride / height and measured stride / height. The figure which shows an example of the relationship between presumed stride / height and measured stride / height. The table | surface which shows an example of the stage of a fall risk. The table | surface which shows the other example of the stage of a fall risk. The block diagram of an example of the health maintenance promotion system using the walk analysis apparatus shown in FIG. The figure which shows an example of the input screen of the personal computer shown in FIG. The figure which shows a part of management ledger before update. The figure which shows a part of management ledger after an update.

Explanation of symbols

10 Accelerometer (acceleration measuring means)
20 arithmetic unit 24 CPU (specific period extracting means, estimated index calculating means, walking ability estimating means)
26 ROM (specific period extraction means, estimation index calculation means, walking ability estimation means)
100 walking analysis device

Claims (7)

Detect vertical acceleration, which is the acceleration in the vertical direction of the waist during walking, longitudinal acceleration, which is the acceleration in the longitudinal direction of the waist during walking, and lateral acceleration, which is the acceleration in the lateral direction of the waist during walking. Acceleration measuring means for measuring the time change of
A period extracting means for extracting a specific period during which a specific walking action is performed based on a time change of at least one of the accelerations;
An estimated index calculating means for calculating an estimated index related to walking ability at the time of walking based on a time change in the specific period of at least one of the accelerations;
Walking ability estimating means for estimating walking ability using the estimated index calculated by the estimated index calculating means and the relationship between the estimated index prepared in advance and the walking ability;
A gait analysis device characterized by comprising:   The period extracting means extracts two time points from the time points when a specific walking motion is performed based on the time change of the acceleration, and extracts between the two time points as the specific time period. The gait analysis apparatus according to claim 1, wherein   The period extracting means extracts the specific period during which a specific walking motion is performed on each of the left and right feet, and the walking ability estimating means estimates the walking ability on the left and right legs. The gait analysis apparatus according to claim 1, wherein the gait analysis apparatus is characterized. Detect vertical acceleration, which is the acceleration in the vertical direction of the waist during walking, longitudinal acceleration, which is the acceleration in the longitudinal direction of the waist during walking, and lateral acceleration, which is the acceleration in the lateral direction of the waist during walking. An acceleration measurement step for measuring the time change of
A period extracting step of extracting a specific period during which a specific walking motion is performed based on a time change of at least one of the accelerations;
An estimated index calculating step for calculating an estimated index related to walking ability at the time of walking based on a temporal change in the specific period of at least one of the accelerations;
A walking ability estimation step for estimating a walking ability using a relationship between the estimated index calculated by the estimated index calculation means and the estimated index prepared in advance and the walking ability;
A gait analysis method characterized by comprising:   The period extracting step extracts two time points from the time points when a specific walking motion is performed based on the time change of the acceleration, and extracts between the two time points as the specific time period. The walking analysis method according to claim 4, wherein the walking analysis method is characterized.   The period extracting step is to extract the specific period during which a specific walking motion is performed on each of the left and right feet, and the walking ability estimating step is to estimate the walking ability of the left and right feet. The gait analysis method according to claim 4, wherein the gait analysis method is characterized.   Information storage means for storing information on the walking ability determined by relying on three-dimensional acceleration using the walking analysis device according to any one of claims 1 to 3, and calculating points based on the information A health maintenance and enhancement system having point calculation means.

Images