JP2021137371A - Walking diagnostic system - Google Patents

Abstract

To provide a walking diagnostic system capable of simply grasping variation in a walking state of a subject.SOLUTION: A walking diagnostic system comprises a measuring apparatus 10 for measuring pressure for each part applied to a sole of a foot over time and a walking diagnostic apparatus 20. The walking diagnostic apparatus 20 comprises: an information processing unit 23 for acquiring a feature amount related to a walking state from measurement information measured by the measuring apparatus 10; a determination unit for determining whether the walking state is an alert state on the basis of the acquired feature amount; and a display unit 25 for notifying that the walking state is the alert state on the basis of a determination result of the determination unit.SELECTED DRAWING: Figure 1

Description

The present invention relates to a gait diagnostic system.

In recent years, musculoskeletal syndrome, which is a condition requiring long-term care due to a disorder of the locomotorium and a condition having a high risk of requiring long-term care, has become a social problem. In order to detect locomotor syndrome at an early stage, technological development related to the diagnosis of locomotor syndrome is underway. For example, in Patent Document 1, the movements of a plurality of predetermined parts of the subject's body are detected from the captured image, the feature parameters are obtained for each part of the subject from the obtained detection information, and the teacher data is learned in advance. A motor function diagnostic device that determines motor function from feature parameters using a neural network is disclosed.

Japanese Unexamined Patent Publication No. 2016-144598

An object of the present invention is to provide a walking diagnosis system capable of easily grasping a change in the walking state of a person to be measured.

The gait diagnosis system that solves the above problems is a gait diagnosis system that diagnoses a walking state, and is a measuring unit that measures the pressure of each part applied to the sole of one foot over time, and a measurement measured by the measuring unit. An information processing unit that acquires a feature amount related to the walking state from information, a determination unit that determines whether or not the walking state is a predetermined alert target state based on the acquired feature amount, and a determination unit. It is provided with a notification unit for notifying that the walking state is a predetermined alert target state based on the determination result of.

According to the above configuration, the person to be measured or another person such as a guardian simply attaches the measurement unit to the person to be measured and walks, and confirms the notification of the alert target state by the walking diagnosis system. It is possible to easily grasp that the walking state of the person to be measured has changed to the state to be alerted without the need for direct diagnosis by a medical worker. Then, when it is notified that the state is the target of the alert, it is possible to take appropriate measures such as having the person to be measured see a medical institution.

In the walking diagnosis system, the determination unit sets the walking state to a predetermined alert target state based on the comparison between the acquired feature amount and the alert information preset for each person to be measured by the medical staff. It is preferable to determine whether or not.

According to the above configuration, alert information can be set according to the walking state of the person to be measured at the time of diagnosis by the medical staff. Therefore, it is possible to suppress the delay in discovering the change to the state of the alert target due to the loose setting of the alert information, and the excessive notification unnecessarily due to the strict setting of the alert information.

In the gait diagnosis system, the information processing unit includes an FTA calculation unit that acquires a femoral tibial angle as the feature amount, and the determination unit is an alert target in which the walking state is a predetermined alert based on the acquired femoral tibial angle. It is preferable to determine whether or not the state is.

The above configuration is effective when it is desired to grasp the state in which the clubfoot and clubfoot have changed as the state to be alerted.
In the walking diagnosis system, the information processing unit includes an LHA calculation unit that acquires a leg heel angle as the feature amount, and the determination unit is an alert target whose walking state is a predetermined alert based on the acquired leg heel angle. It is preferable to determine whether or not the state is.

The above configuration is effective when it is desired to grasp the state in which the clubfoot and clubfoot have changed as the state to be alerted.
In the gait diagnosis system, the information processing unit divides a step section from the landing of the foot to the takeoff during walking into a plurality of subsections, and the foot pressure of the sole of the foot at regular intervals in the subsection. When the coordinate calculation unit that calculates the two-dimensional coordinates of the center and the three consecutive two-dimensional coordinates at regular time intervals in the same subsection are sequentially set to coordinates P _n , coordinates P _{n + 1} , and coordinates P _{n + 2} . deflection value calculation for calculating a vector coordinate deflection value is the sum of the deflection angle formed by the vector indicating the amount of change from P _{n + 1} to the coordinate P _{n + 2,} which shows the amount of change from the coordinate P _n to the coordinate P _{n + 1} as the feature quantity It is preferable that the determination unit determines whether or not the walking state is a predetermined alert target state based on the acquired runout value.

The runout value is a value that quantifies how much the center of foot pressure swings at each timing of one step from landing to takeoff. Therefore, by determining the determination unit based on the runout value, it is possible to grasp the change in the walking state accompanied by the change in the smoothness of one step from landing to takeoff.

According to the present invention, it is possible to easily grasp that the walking state has changed to a predetermined alert state.

Explanatory drawing of the gait diagnostic equipment system. Explanatory drawing of arrangement of a pressure sensor. Explanatory diagram of feature data. Explanatory drawing of information processing department. Explanatory drawing of judgment processing part. Graph of waveform data. Graph of waveform data of one step section. A graph of two-dimensional coordinates of the center of foot pressure. Explanatory drawing of the runout angle. A graph of waveform data related to feature data classified into group 0 and a graph of two-dimensional coordinates of the center of foot pressure. A graph of waveform data related to feature data classified into group 1 and a graph of two-dimensional coordinates of the center of foot pressure. A graph of waveform data related to feature data classified into group 2 and a graph of two-dimensional coordinates of the center of foot pressure. Flow chart of gait diagnosis.

Hereinafter, the walking diagnosis system of the present embodiment will be described.
As shown in FIG. 1, the walking diagnosis system includes a measuring device 10 as a measuring unit and a walking diagnosis device 20.

As shown in FIGS. 1 and 2, the measuring device 10 includes a base 11 used as an insole for shoes. Four pressure sensors, a heel sensor 12a, a toe sensor 12b, an inner sensor 12c, and an outer sensor 12d, are attached to the base 11 as pressure sensors 12 for detecting the pressure of each part applied to the sole of the foot during walking. There is.

As shown in FIG. 2, at the base 11, the heel sensor 12a is arranged at a portion where the load of the heel is applied, and detects the pressure applied to the heel of the sole of the foot. The toe sensor 12b is arranged at any of the toe-loaded portions of the 2nd to 4th toes and detects the pressure applied to the toes of the sole of the foot.

The inner sensor 12c is located inside the line L connecting the heel sensor 12a and the toe sensor 12b and is arranged at a portion where the load of the ball of the finger is applied, and detects the pressure applied to the inner portion of the sole of the foot. The outer sensor 12d is located outside the line L and is located at the portion where the load of the hypothenar eminence is applied, and detects the pressure applied to the outer portion of the sole of the foot.

In other words, the heel sensor 12a and the toe sensor 12b are arranged so as to detect the pressure at the first position and the second position which are separated from each other in the front-rear direction on the sole of the foot. The inner sensor 12c and the outer sensor 12d are arranged so as to detect the pressure at the third position and the fourth position separated in the left-right direction across the line L connecting the first position and the second position.

Each pressure sensor 12 independently detects the pressure applied to the sole of the foot at predetermined time intervals. The predetermined time is, for example, 5 to 30 milliseconds. In this embodiment, the pressure is set to be detected every 20 milliseconds.

As the pressure sensor 12, a known pressure sensor using a piezoelectric element or the like can be used. In particular, in view of the usage situation in which the sensor is arranged on the sole of the foot, it is preferable to use an elastomer-made capacitance type sensor using a dielectric elastomer from the viewpoint of elasticity and durability. Examples of the dielectric elastomer include crosslinked polyrotaxane, silicone elastomer, acrylic elastomer, and urethane elastomer.

As shown in FIG. 1, the measuring device 10 is equipped with a transmitting unit 13 that transmits each detected value detected by each pressure sensor 12 to the walking diagnostic device 20. Each detected value detected by each pressure sensor 12 is a detected value that can be converted into a pressure value according to the detection method of the pressure sensor such as a capacitance value and an electric resistance value. Therefore, the "detected value" described below can be read as the measured value of the pressure at the position where the pressure sensor 12 on the sole of the foot is attached.

The measuring device 10 is prepared for both the left foot and the right foot, and if necessary, only one of the left and right or both left and right can be used.
As shown in FIG. 1, the walking diagnosis device 20 is composed of a receiving unit 21 that receives measurement information transmitted from a transmitting unit 13 of the measuring device 10, a storage unit 22 that stores the received measurement information, and the measurement information. It is provided with an information processing unit 23 that acquires a feature amount related to the walking state. The walking diagnosis device 20 further includes a determination processing unit 24 that determines the walking state based on the feature amount acquired by the information processing unit 23, a display unit 25 that displays the determination result of the determination processing unit 24, and various kinds of information. It is provided with an input unit 26 for inputting.

As the walking diagnosis device 20, for example, a computer such as a mobile terminal or a tablet terminal can be used. As the display unit 25, for example, a known display device such as a liquid crystal display can be used. As the input unit 26, for example, a known input device such as a keyboard, a mouse, or a touch panel can be used. The receiving unit 21 of the walking diagnosis device 20 has a wired or wireless communication means, and communicates with the transmitting unit 13 of the measuring device 10 by a known communication method.

The storage unit 22 is configured to store the measurement information transmitted from the measuring device 10 and to store the feature amount data based on the measurement information. As shown in FIG. 3, the feature amount data includes identification information such as a date and time and a user name, and feature amounts F1 to F13 related to a walking state.

The storage unit 22 stores a plurality of feature data such as past feature data of the same subject, feature data of different subjects, and model data set by an expert such as a medical worker. Examples of the model data include feature data obtained from measurement results of subjects having specific symptoms such as clubfoot and clubfoot.

Further, the storage unit 22 stores alert information regarding the feature amount. Details of the alert information will be described later. Further, the storage unit 22 stores a program for causing the computer to execute each process in the information processing unit 23 and the determination processing unit 24.

Examples of the storage unit 22 include HDDs, SSDs, and semiconductor memory elements. Further, the storage unit 22 may be a storage device connected to the walking diagnosis device 20 via a network.

As shown in FIG. 4, the information processing unit 23 includes a waveform data creation unit 31, a division unit 32, a section maximum value acquisition unit 33, a section time acquisition unit 34, a coordinate calculation unit 35, a stay ratio calculation unit 36, and a runout value calculation. A unit 37, a classification unit 38, an FTA calculation unit 39, an LHA calculation unit 40, and a first display processing unit 41 are provided.

The waveform data creation unit 31 is waveform data representing a time change of a detected value for each part of the sole of one foot detected by each pressure sensor 12 based on the measurement information to be analyzed stored in the storage unit 22. To create. FIG. 6 shows an example of waveform data created by the waveform data creation unit 31. In FIG. 6, the solid line indicates the detection value of the heel sensor 12a, the alternate long and short dash line indicates the detection value of the toe sensor 12b, the alternate long and short dash line indicates the detection value of the inner sensor 12c, and the broken line indicates the detection value of the outer sensor 12d. Indicates the detected value.

From the created waveform data, the division unit 32 extracts a section for a specific step from the landing of the foot during walking to the takeoff as a target section, and further divides the extracted target section into a plurality of subsections. ..

As shown in FIG. 6, the waveform data during walking is a waveform that periodically repeats a section in which the pressure value is substantially constant and a section in which the detected value changes. In the waveform data during walking, the section where the detected value is substantially constant is the section where the foot is away from the ground, and the section where the detected value is changed is the section where the foot is in contact with the ground. The dividing portion 32 has a starting point t1 at which the heel pressure value starts to rise from a section where the detected value of each part is substantially constant, and is the most of the starting points where the detected value of each part is substantially constant. A specific step section A to be diagnosed is extracted from the step section A having the late start point as the end point t2.

The one-step section A to be extracted is selected by the division unit 32 based on a preset standard. As the above criteria, for example, the step section A located in the preset order such as the 10th step, the 20th step, etc. from the start of measurement is selected, and the step is one step after a certain period of time has passed from the previously extracted step section A. Examples include selecting the one-step section A, dividing the waveform data by a fixed number of steps or time, and randomly selecting the one-step section A within each of the divided sections.

As shown in FIG. 7, the dividing section 32 divides the extracted step section A into a small section of a heel section Ah, a toe section At, and an intermediate section Am.
The heel section Ah is an initial small section of the step section A in which pressure is relatively applied to the heel side. In the present embodiment, the section from the start point t1 of the step section A to the time point t3 where the detection value of the heel sensor 12a becomes the apex of the peak is defined as the heel section Ah.

The toe section At is a small section at the end of the one-step section A in which pressure is relatively applied to the toe side. In the present embodiment, the section from the time point t4 where the detected value of the toe sensor 12b is the largest among the detected values of the four pressure sensors 12 to the end point t2 of the step section A is defined as the toe section At. Depending on how you walk, there may be no toe section At.

The intermediate section Am is a small section excluding the heel section Ah and the toe section At from the step section A. If there is a toe section At, the section from the time point t3 to the time point t4 is set as the intermediate section Am, and if there is no toe section At, the section from the time point t3 to the time point t2 is set as the intermediate section Am.

The section maximum value acquisition unit 33 includes the maximum value of the detected values of the four pressure sensors 12 in the heel section Ah (hereinafter, referred to as the heel section maximum value Ah _max ), and the four pressure sensors 12 in the toe section At. The maximum value of the detected value (hereinafter, referred to as the toe section maximum value At _max ) is acquired from the measurement information. Then, the acquired heel section maximum value Ah _max and the toe section maximum value At _max are stored in the storage unit 22 as the feature amounts F1 and F2 constituting the feature amount data shown in FIG.

The section time acquisition unit 34 acquires the section time Th, which is the length of the heel section Ah, the section time Tm, which is the length of the intermediate section Am, and the section time Tt, which is the length of the toe section At, from the measurement information. Then, the acquired section times Th, Tm, and Tt are stored in the storage unit 22 as feature quantities F3 to F5 constituting the feature quantity data shown in FIG. In addition, each section time may be the time itself such as the number of seconds, or may be a parameter corresponding to the section time such as the number of measurement points in the corresponding section.

As shown in FIG. 8, the coordinate calculation unit 35 has two dimensions of the foot pressure center for each detection time in each subsection of the heel section Ah, the toe section At, and the intermediate section Am based on the measurement information of the step section A. The coordinates (X (t), Y (t)) are calculated. X (t) is the left-right coordinate of the foot pressure center at time t, and Y (t) is the front-back coordinate of the foot pressure center at time t. The two-dimensional coordinates of the center of foot pressure can be obtained from the detected values of the four pressure sensors 12 detected at the same detection time. The points 12a to 12d in the two-dimensional coordinates shown in FIG. 8 indicate the positions of the pressure sensors 12 having the corresponding symbols.

As shown in FIG. 8, the stay ratio calculation unit 36 divides the two-dimensional coordinate system at the center of foot pressure into three preset regions B1, B2, and B3. Then, the stay ratio calculation unit 36 has a stay ratio Rh1 and a time to stay in the area B2, which is a ratio of the time to stay in the area B1 in the heel section Ah, based on the two-dimensional coordinates calculated by the coordinate calculation unit 35. The stay ratio Rh2, which is the ratio of the above, and the stay ratio Rh3, which is the ratio of the time spent in the area B3, are calculated. Then, the calculated stay ratios Rh1, Rh2, and Rh3 are stored in the storage unit 22 as the feature amount F6 constituting the feature amount data shown in FIG.

Similarly, in the intermediate section Am, the stay ratio calculation unit 36 has a stay ratio Rm1 which is a ratio of the time spent in the area B1, a stay ratio Rm2 which is a ratio of the time stayed in the area B2, and a time spent in the area B3. The stay ratio Rm3, which is the ratio of Then, the calculated stay ratios Rm1, Rm2, and Rm3 are stored in the storage unit 22 as the feature amount F7 constituting the feature amount data shown in FIG.

Similarly, in the toe section At, the stay ratio calculation unit 36 has a stay ratio Rt1 which is a ratio of the time spent in the area B1, a stay ratio Rt2 which is a ratio of the time stayed in the area B2, and a time spent in the area B3. The stay ratio Rt3, which is the ratio of, is calculated. Then, the calculated stay ratios Rt1, Rt2, and Rt3 are stored in the storage unit 22 as the feature amount F8 constituting the feature amount data shown in FIG.

Each of the above stay ratios may be a numerical value based on the stay time itself such as the number of seconds spent in the corresponding area, or corresponds to the stay time such as the number of measurement points located in the area in the two-dimensional coordinate system. It may be a numerical value based on the parameters to be used.

Based on the two-dimensional coordinates calculated by the coordinate calculation unit 35, the runout value calculation unit 37 sets the runout value Sh of the heel section Ah and the runout value Sm of the intermediate section Am as runout values indicating the runout of the center of foot pressure. The runout value St of the toe section At is calculated. The runout value Sh of the heel section Ah is calculated as follows.

As shown in FIG. 9, in the heel section Ah, the two-dimensional coordinates of the foot pressure center at three consecutive measurement points are coordinate P _n , coordinate P _{n + 1} , and coordinate P _{n + 2} , respectively, and coordinates P n to coordinates P _n. The angle formed by the vector V1 indicating the amount of change to _{n + 1} and the vector V2 indicating the amount of change from the coordinate P _{n + 1 to the coordinate P n + 2} is defined as the deflection angle θ.

The runout value calculation unit 37 calculates the runout angle θ when the _{above coordinates P n} are used for the two-dimensional coordinates of all foot pressure centers except the last two in the heel section Ah, and these runout angles θ. Is added up to calculate the runout value Sh, which is the sum of the runout angles θ. The runout value Sm of the intermediate section Am and the runout value St of the toe section At are also calculated in the same manner. Then, the runout value calculation unit 37 stores the calculated runout values Sh, Sm, and St in the storage unit 22 as feature amounts F9 to F11 constituting the feature amount data shown in FIG.

The classification unit 38 includes 11 feature quantities constituting the feature quantity data by targeting a plurality of sets of feature quantity data stored in the storage unit 22, including the feature quantity data consisting of newly acquired feature quantities. The target group is clustered into a plurality of subsets based on F1 to F11. Clustering is performed by dimensionally compressing feature data by principal component analysis and using a known method such as the k-means method or the GMM method.

FIGS. 10 to 12 show an example of the result of clustering a plurality of feature data by the k-means method. Here, clustering produces three subsets of group 0, group 1, and group 2. Then, three sets of waveform data of one step section A and two-dimensional coordinate data of the center of foot pressure, which are the sources of the feature data classified into each group, are shown in FIGS. 10 to 12. Figures 10-12 show the raw data after the measurement.

Group 0 shown in FIG. 10 is a group that can be labeled as having an average walking style. As a feature of group 0, the heel section Ah, the intermediate section Am, and the toe section At are clearly present in the step section A, and the load is transferred from the heel to the toe via the central part of the sole.

Group 1 shown in FIG. 11 is a group that can be labeled as a walking style that tends to walk on the outside. As a feature of Group 1, most of the step section A is occupied by the heel section Ah and the intermediate section Am, and the kicking motion with the toes is not included.

Group 2 shown in FIG. 12 is a group that can be labeled as a walking style that tends to walk inward. As a feature of Group 2, most of the step section A is occupied by the heel section Ah and the toe section At, and the toes arrive early after the heel reaches the ground.

The FTA calculation unit 39 calculates the femoral tibial angle (hereinafter referred to as FTA) of the subject based on the measurement information of the step section A. Then, the FTA calculation unit 39 stores the calculated FTA in the storage unit 22 as the feature amount F12 constituting the feature amount data shown in FIG.

The FTA is the angle formed by the long axis of the femur and the long axis of the tibia, and is used as one of the criteria for determining clubfoot and clubfoot. Since the FTA is a parameter indicating the state of the knee joint, it has a close relationship with the walking state, and the pressure distribution on the sole of the foot, specifically, the heel sensor 12a, the toe sensor 12b, the medial sensor 12c, and the like. It can be estimated from the balance of the load detected by the outer sensor 12d. A specific FTA estimation method will be described later.

The LHA calculation unit 40 calculates the leg heel angle (hereinafter referred to as LHA) of the person to be measured based on the measurement information of the step section A. Then, the LHA calculation unit 40 stores the calculated LHA in the storage unit 22 as the feature amount F13 constituting the feature amount data shown in FIG.

LHA is the angle formed by the long axis of the lower leg and the axis of the calcaneus, and is used as one of the criteria for determining clubfoot and clubfoot. Since LHA is a parameter indicating the inclination of the foot and the heel, it has a close relationship with the walking state, and the pressure distribution on the sole of the foot, specifically, the heel sensor 12a, the toe sensor 12b, and the medial sensor 12c. , And the load balance detected by the outer sensor 12d. A specific method for estimating LHA will be described later.

The first display processing unit 41 includes the feature amount data shown in FIG. 3, the waveform data of the step section A shown in FIG. 7, the two-dimensional coordinate data of the foot pressure center shown in FIG. 7, and the feature amount data classified by the classification unit 38. Create an image or the like for displaying the group or the like on the display unit 25. Then, the first display processing unit 41 causes the display unit 25 to display the created image or the like when a predetermined operation is performed through the input unit 26.

An example of a method for estimating LHA and FTA is a method using an angle estimation model. In the method using the angle estimation model, first, teacher data including the actual measurement results of LHA and FTA, basic data, and walking data is collected. Basic data consists of biometric information such as gender, age, height and weight. The walking data includes walking information measured by the pressure sensor 12, such as left-right balance, front-back balance, takeoff / landing acceleration, takeoff / landing strength (load value), and swing width. Next, by machine learning the collected teacher data with a neural network, an angle estimation model that estimates and outputs the measurement results of LHA and FTA is constructed with respect to the input basic data and walking data. By using such a trained angle estimation model, LHA and FTA can be estimated based on the measurement information of the subject. The swing width, which is the size of the swing of the body, is closely related to LHA and FTA, which are affected by changes in the state of the locomotorium. Therefore, the estimation accuracy of the angle estimation model can be improved by learning the swing width as walking information.

As shown in FIG. 5, the determination processing unit 24 includes a first determination unit 51, a second determination unit 52, a third determination unit 53, and a second display processing unit 54. The determination processing unit 24 sets the walking state of the person to be measured as a predetermined alert target based on the comparison between the feature amount acquired by the information processing unit 23 and the alert information regarding the feature amount stored in the storage unit 22. Judge whether or not it is in the state of. Examples of the alert target state include a state for prompting an examination and a medical examination, a state for calling attention, and a state for prompting a break. Hereinafter, the status of the alert target is described as the alert status.

The storage unit 22 stores the FTA alert range, which is alert information related to FTA, and the LHA alert range, which is alert information related to LHA, as alert information. The alert information is information that serves as a reference when determining whether or not the walking state of the person to be measured is in the alert state by using the above-mentioned feature amount.

The alert information is set for each person to be measured by the medical staff based on the latest diagnosis result by the medical staff for the person to be measured, and is stored in the storage unit 22 through the input unit 26. The setting by the medical staff includes the case where the setting is made by another person such as a person to be measured, a guardian, a caregiver, etc. based on the instruction of the medical staff.

The FTA alert range is set, for example, as a range that changes beyond the determination angle with respect to the FTA of the person to be measured who has been diagnosed most recently. The determination angle is, for example, 5 to 10 degrees. As an example, when the person to be measured has a tendency to clubfoot and the diagnosis value of the FTA diagnosed most recently by the person to be measured is 176 degrees, the change is made in the direction of increasing the judgment angle of 5 degrees or more from the diagnosis value. The range of 181 degrees or more, which is the range, is set as the FTA alert range. In addition, when the person to be measured has a tendency to valgus and the diagnosis value of the FTA diagnosed most recently by the person to be measured is 170 degrees, the range changed from the diagnosis value in the direction of decreasing by 5 degrees or more, which is the judgment angle. The range of 165 degrees or less is set as the FTA alert range.

The LHA alert range is set, for example, as a range in which the LHA of the person to be diagnosed most recently changed to a determination angle or more. The determination angle is, for example, 5 to 10 degrees. As an example, when the subject has a tendency to valgus (pronation) and the Achilles tendon is stiff, the Achilles tendon is likely to be damaged. From the diagnosis value of LHA diagnosed most recently of the subject who has this tendency, the LHA alert range is set at the place where the valgus angle increases by a certain amount in consideration of the activity amount of the subject.

The first determination unit 51 determines whether or not the calculated value of the FTA calculated by the FTA calculation unit 39 of the information processing unit 23 is included in the FTA alert range. Then, when the calculated value of the FTA is included in the LHA alert range, the first determination unit 51 outputs an FTA alert signal to the second display processing unit 54.

The second determination unit 52 determines whether or not the calculated value of LHA calculated by the LHA calculation unit 40 of the information processing unit 23 is included in the LHA alert range. Then, when the calculated value of LHA is included in the LHA alert range, the second determination unit 52 outputs the LHA alert signal to the second display processing unit 54.

The third determination unit 53 determines whether or not the classified subset as a result of clustering by the classification unit 38 of the information processing unit 23 is the same as the classified subset of the feature amount data of the one-step interval A extracted last time. To judge. Then, when the subset classified this time is different from the subset classified last time, the third determination unit 53 outputs the classification alert signal to the second display processing unit 54.

The fact that the classified subsets are different from the previous one means that the walking state has changed abruptly. Therefore, in the determination of the third determination unit 53, the state in which the walking state suddenly changes corresponds to the alert state, and by using the result of clustering, it is determined that the walking state has changed suddenly.

In the present embodiment, the first determination unit 51, the second determination unit 52, and the third determination unit 53 determine whether or not the walking state is a predetermined alert target state based on the acquired feature amount. Corresponds to the judgment unit to be used.

The second display processing unit 54 is a warning message based on the FTA alert signal output from the first determination unit 51, the LHA alert signal output from the second determination unit 52, and the classification alert signal output from the third determination unit 53. Is displayed on the display unit 25. For example, when an FTA alert signal is input, a warning message such as "FTA (181 degrees) is within the FTA alert range, so please consult a medical institution." Is displayed on the display unit 25. In the present embodiment, the display unit 25 corresponds to a notification unit that notifies that the walking state is a predetermined alert target state.

Next, a gait diagnosis using the gait diagnosis system of the present embodiment will be described.
First, the medical staff diagnoses the current walking condition of the person to be measured. Based on the diagnosis result, the medical staff sets the walking state to be noted in the future such as worsening clubfoot tendency to the alert state, and stores the FTA alert range and the LHA alert range corresponding to the alert state in the storage unit 22. To memorize.

After that, as a measurement step, the person to be measured wears a shoe in which the measuring device 10 is arranged on the sole of the shoe, and wears a mobile terminal as the walking diagnostic device 20. Perform life movements. Here, a case where only the left foot is the measurement target, that is, a case where the measuring device 10 is arranged only on the sole of the shoe for the left foot will be described.

The measuring device 10 detects the pressure of each part applied to the sole of the left foot of the person to be measured by each pressure sensor 12 during daily activities, and uses the detected detection value as measurement information in the walking diagnostic device 20 at each detection timing. Send to. The detection of the pressure applied to the sole of the foot by the measuring device 10 and the transmission of the measurement information to the gait diagnostic device 20 are always executed while wearing the shoes on which the measuring device 10 is arranged.

The measurement information transmitted from the measuring device 10 is stored in the storage unit 22 of the walking diagnosis device 20. The gait diagnosis device 20 performs gait diagnosis of the person to be measured based on the measurement information stored in the storage unit 22.

The details of the walking diagnosis by the walking diagnosis device 20 will be described below based on the flowchart shown in FIG.
When new measurement information is stored in the storage unit 22, the waveform data creation unit 31 creates waveform data based on the measurement information stored in the storage unit 22 as the waveform creation step S11.

Subsequently, as the extraction step S12, the division unit 32 extracts a specific step section A from the created waveform data, and as the division step S13, the step section A is set as the heel section Ah, the toe section At, and the intermediate section Am. Divide into small sections. The steps after the division step S13 are executed in the step section A unit extracted by the extraction step S12, and are repeatedly executed for all the extracted step sections A.

Subsequently, as the section maximum value acquisition step S14, the section maximum value acquisition unit 33 acquires the heel section maximum value Ah _max and the toe section maximum value At _max in the step section A, and constitutes a new feature amount data. It is stored in the storage unit 22 as F1 and F2.

Subsequently, as the section time acquisition step S15, the section time acquisition unit 34 acquires the section time Th of the heel section Ah, the section time Tm of the intermediate section Am, and the section time Tt of the toe section At in the step section A, and newly obtains the section time Tt. The feature amounts F3 to F5 that constitute the feature amount data are stored in the storage unit 22.

Subsequently, in the coordinate calculation step S16, the coordinate calculation unit 35 calculates the two-dimensional coordinates of the foot pressure center in the step section A. After that, as the stay ratio calculation step S17, the stay ratio calculation unit 36 sets the stay ratio Rh1, Rh2, Rh3 of the foot pressure center of the heel section Ah and the intermediate section Am based on the calculated two-dimensional coordinates of the foot pressure center. The stay ratio Rm1, Rm2, Rm3 at the center of the foot pressure and the stay ratio Rt1, Rt2, Rt3 at the center of the foot pressure in the toe section At are acquired, and the feature amounts F6 to F8 constituting the new feature amount data are stored in the storage unit 22. Remember.

Subsequently, in the runout value calculation step S18, the runout value calculation unit 37 calculates the runout values Sh, Sm, and St, and stores them in the storage unit 22 as feature amounts F9 to F11 constituting new feature amount data.

Subsequently, as the classification step S19, the classification unit 38 performs clustering using the new feature amount data and the feature amount data stored in the storage unit 22 as a target group, and classifies the feature amount data into a plurality of groups. .. Further, the classification unit 38 stores the clustering result in the storage unit 22.

Subsequently, in the FTA calculation step S20, the FTA calculation unit 39 calculates the FTA and stores the calculated value of the FTA in the storage unit 22. Subsequently, in the LHA calculation step S21, the LHA calculation unit 40 calculates the LHA and stores the calculated value of the LHA in the storage unit 22.

Subsequently, as the FTA determination step S22, the first determination unit 51 determines whether or not the calculated value of the FTA calculated by the FTA calculation unit 39 is included in the FTA alert range, and the calculated value of the FTA is the LHA alert range. When included in, an FTA alert signal is output to the second display processing unit 54.

Subsequently, as the LHA determination step S23, the second determination unit 52 determines whether or not the calculated value of the FTA calculated by the LHA calculation unit 40 is included in the FTA alert range, and the calculated value of the LHA is the LHA alert range. When included in, an LHA alert signal is output to the second display processing unit 54.

Subsequently, as the classification determination step S24, the third determination unit 53 classifies the feature amount data of the one-step section A previously extracted by the classified subset as a result of clustering by the classification unit 38 of the information processing unit 23. Determine if it is the same as the subset. Then, when the subset classified this time is different from the subset classified last time, the third determination unit 53 outputs the classification alert signal to the second display processing unit 54.

Subsequently, as the notification step S25, when any one of the FTA alert signal, the LHA alert signal, and the classification alert signal is output, the second display processing unit 54 responds to the output alert signal. By displaying the warning text on the display unit 25, it is notified that the walking state is the alert state. In the present embodiment, the feature amount acquisition step is each step of S11 to S21, and the determination step is each step of S22 to S24.

Next, the operation of this embodiment will be described.
In the walking diagnosis system of the present embodiment, a feature amount related to the walking state is acquired from the measurement information measured during walking, and based on the acquired feature amount, it is determined whether or not the walking state is an alert state. Will be done. When it is determined that the patient is in the alert state, the display unit 25 of the mobile terminal worn by the person to be measured displays that the walking state is in the alert state, and the display indicates that the medical institution is in the alert state. You will be encouraged to see a doctor.

As described above, according to the walking diagnosis system of the present embodiment, it is possible to diagnose whether or not the walking state of the person to be measured is an alert state for walking in activities of daily living. Therefore, the person to be measured can grasp at an early stage that the walking state is an alert state regardless of the presence or absence of subjective symptoms, and can receive detailed diagnosis and treatment by a medical institution. Further, the information such as the feature amount data stored in the gait diagnostic device 20 helps the medical staff to appropriately grasp the symptom of the person to be measured. In particular, it is useful when the person to be measured has subjective symptoms such as discomfort, but the symptoms cannot be verbalized.

Next, the effect of this embodiment will be described.
(1) The gait diagnosis system includes a measuring device 10 for measuring the pressure applied to the sole of one foot over time, and a gait diagnosis device 20. The walking diagnosis device 20 has an information processing unit 23 that acquires a feature amount related to the walking state from the measurement information measured by the measuring device 10, and whether or not the walking state is an alert state based on the acquired feature amount. It is provided with a determination unit for determining whether or not, and a display unit 25 for notifying that the walking state is an alert state based on the determination result of the determination unit.

According to the above configuration, the person to be measured walks with the measuring device 10 attached and only confirms the notification that the alert state is in the state, and the person to be measured does not require a direct diagnosis by a medical worker. It is possible to easily grasp that the walking state of the person has changed to the alert state. Then, when the alert state is notified, the person to be measured can take appropriate measures such as visiting a medical institution. In addition, the gait diagnosis system having the above configuration enables gait diagnosis for walking in activities of daily living, and can grasp at an early stage that the gait state of the person to be measured has changed to an alert state.

(2) The first determination unit 51 and the second determination unit 52 determine whether or not the walking state is the alert state based on the comparison between the acquired feature amount and the alert information. The alert information is set for each person to be measured by the medical staff based on the diagnosis result of the medical staff.

According to the above configuration, alert information can be set according to the walking state of the person to be measured at the time of diagnosis by the medical staff. Therefore, it is possible to suppress the delay in discovering the change to the alert state due to the loosely set alert information, and the unnecessary excessive notification due to the strict setting of the alert information.

(3) The first determination unit 51 determines whether or not the walking state is an alert state based on the FTA acquired as a feature amount. The second determination unit 52 determines whether or not the walking state is an alert state based on the LHA acquired as the feature amount.

The above configuration is effective when it is desired to grasp the changed state of clubfoot and clubfoot as an alert state.
(4) In the third determination unit 53, the walking state is in the alert state based on whether or not the classified subset is the same as the classified subset in the feature amount data of the one-step interval A extracted last time. Determine if it exists. That is, it is determined whether or not the walking state is the alert state based on the comparison of the feature amounts acquired from the same subject at different timings.

According to the above configuration, it is possible to grasp a sudden change in the walking state.
(5) The information processing unit 23 divides the step section from the landing of the foot during walking to the takeoff into a plurality of small sections Ah, Am, and At, and the foot sole in the step section A at regular intervals. It is provided with a coordinate calculation unit 35 for calculating the two-dimensional coordinates of the foot pressure center of the foot pressure, and a runout value calculation unit 37 for calculating the runout values Sh, Sm, and St in small section units. The third determination unit 53 determines whether or not the walking state is the alert target based on the runout values Sh, Sm, and St.

The runout values Sh, Sm, and St are values that quantify how much the center of foot pressure swings at each timing of one step from landing to takeoff. Therefore, by performing the determination of the determination unit based on the runout values Sh, Sm, and St, it is possible to grasp the change in the walking state accompanied by the change in the smoothness of one step from landing to takeoff.

In addition, when a medical worker examines a person to be measured, by checking the runout values Sh, Sm, and St, the smoothness of one step from landing to takeoff can be quantitatively judged, and the person to be measured can be measured. Can properly grasp the symptoms of a person. In particular, by setting the runout values Sh, Sm, and St subdivided into subdivision units in which the one-step section A is divided, it is possible to perform a more detailed analysis of when the center of foot pressure is swinging.

In addition, the swing of the center of foot pressure in each subsection tends to change according to the movement of the foot below the knee during walking. For example, when the subject has a tendency to walk outside or inside, it is easy to obtain characteristic runout values Sh, Sm, and St. Therefore, when the medical staff examines the person to be measured, the walking state of the person to be measured can be grasped in more detail by confirming the runout values Sh, Sm, and St in small section units obtained by dividing the section A one step. ..

(6) Based on the relative relationship of the pressure of each part applied to the sole of the foot, the divided portion 32 includes the one-step section A, the heel section Ah in which the pressure is applied to the heel side, and the toe section At in which the pressure is applied to the toe side. And the intermediate section Am excluding the heel section Ah and the toe section At in the one-step section A.

According to the above configuration, the heel section Ah in which pressure is applied to the heel side and the toe section At in which pressure is applied to the toe side can be set more accurately. Therefore, the accuracy of each feature amount such as the runout value obtained based on the divided subsections is improved.

(7) The measurement information measured by the measuring device 10 includes the first position and the second position separated in the front-rear direction on the sole of the foot, and the line L connecting the first position and the second position in the left-right direction. Includes detected values obtained by detecting the pressure of each of the separated third and fourth positions over time.

According to the above configuration, the pressure distribution on the sole of the foot and the two-dimensional coordinates of the center of the foot pressure can be obtained more accurately. As a result, the accuracy of the FTA, LHA, and runout values Sh, Sm, and St acquired from the measurement information is improved.

(8) Stay ratio calculation in which the coordinate system of the two-dimensional coordinates of the foot pressure center is divided into a plurality of regions B1 to B3, and the stay ratio of the two-dimensional coordinates of the foot pressure center in each region B1 to B3 is calculated in small section units. The unit 36 is provided.

The stay ratio of the two-dimensional coordinates of the foot pressure center in small section units obtained by the above configuration can be regarded as a value indicating the direction in which the foot pressure center swings at each timing of one step from landing to takeoff. Therefore, by performing the determination of the determination unit based on the stay ratio, it is possible to grasp the change in the walking state accompanying the change in the direction in which the center of foot pressure swings. In addition, when the medical staff examines the person to be measured, the walking state of the person to be measured can be grasped in more detail by confirming the above-mentioned stay ratio together with the fluctuation value in units of small sections.

In addition, this embodiment can be implemented by changing as follows. The present embodiment and the following modified examples can be implemented in combination with each other within a technically consistent range.
-In the above embodiment, the case where only the left foot is the measurement target has been described, but only the right foot may be the measurement target, or both the left and right feet may be the measurement target.

-Measurement steps are not limited to those intended for walking in activities of daily living. For example, as a measurement step, a regular walking test such as once a day may be performed. As a walking test, for example, walking on a flat surface for about 10 m can be mentioned.

-The timing of performing the walking diagnosis by the walking diagnosis device 20 may be changed. For example, in the above embodiment, the measurement step of performing the measurement using the measuring device 10 and the walking diagnosis by the walking diagnosis device 20 are simultaneously advanced, but after the measurement step is completed, the walking by the walking diagnosis device 20 is performed. A diagnosis may be made.

The number and arrangement of the pressure sensors 12 provided in the measuring device 10 are not limited to the configuration of the above embodiment, and may be appropriately changed according to the type of feature amount to be acquired. When FTA and LHA are acquired as feature quantities, they are separated in the left-right direction across the first and second positions on the sole of the foot, and the line L connecting the first and second positions. It is preferable to provide the pressure sensors 12 at the third position and the fourth position.

The measuring device 10 may include a storage unit that stores measurement information. When a removable recording medium such as a memory card is used as the storage unit of the measuring device 10, the transmitting unit 13 of the measuring device 10 and the receiving unit 21 of the walking diagnosis device 20 may be omitted.

-The type of feature amount used for the determination in the determination unit and the combination thereof are not limited to the configuration of the above embodiment, and can be appropriately selected from the various feature amounts acquired in the above embodiment. For example, instead of FTA and LHA, even if the alert state is determined by comparing the feature amount of any of the feature amounts F1 to F11 used for clustering by the classification unit 38 with the alert information related to the feature amount. good.

Further, the feature amount used for the determination in the determination unit may be a feature amount other than the feature amount acquired in the above embodiment. Other feature quantities include, for example, walking speed, synchronization of left and right feet during walking, continuous walking time, continuous stopping time such as sitting time, and the like.

-The alert information used for the judgment of the judgment unit is not limited to the value or range set for each person to be measured by the medical staff based on the diagnosis result of the medical staff. For example, the alert information may be a value or range set by the person to be measured, a guardian, a caregiver, etc. based on self-diagnosis, or a value or range common to each classification according to gender, age, etc. There may be.

-The alert information used for the judgment of the determination unit is not limited to the value or range of the feature amount itself, but is the feature amount of the previous time or the predetermined time before which is acquired from the same subject and stored in the storage unit 22. It may be the value or range of the rate of change from.

-The determination unit may be configured to determine whether or not it is in the alert state using AI from the acquired feature amount. Judgment using AI is based on the feature amount that can change rapidly during a series of walking movements such as walking speed, synchronization of left and right feet during walking, continuous walking time, continuous stopping time, and the like. This is especially useful when grasping a suddenly changed state as an alert state. In this case, the AI is set to a normal state by learning one or more types of features before the previous time, which are acquired from the same subject and stored in the storage unit 22, and acquired this time. When the feature amount is not included in the normal state, it is determined as an alert state.

-In the above embodiment, the feature amount acquisition step and the determination step after the division step S13 are performed for the specific step interval A extracted from the waveform data created from the measurement information, but they are included in the waveform data. The feature amount acquisition step and the determination step may be performed for all the step sections A.

-The notification unit that notifies that the walking state is the alert target is not limited to the display unit 25 that displays the warning text. For example, it may be a notification unit that notifies that the walking state is an alert target by generating sound, voice, vibration, light, and a combination thereof. Further, the notification unit may be provided in the measuring device 10.

-The timing of notification may be changed by the notification unit. For example, as a configuration in which notification is performed when the one-step section A in which the alert signal is output continues for a predetermined number of times, or when the ratio of the one-step section A in which the alert signal is output exceeds a predetermined value within a certain period. May be good.

-The notification unit may notify others other than the person to be measured that the walking state of the person to be measured is an alert state. In this case, a device such as a mobile terminal owned by a person other than the person to be measured may be used as the notification unit. Examples of others other than the person to be measured include parents, caregivers, caregivers, and medical staff.

The method of setting the start point t1 and the end point t2 of the one-step section A is not limited to the above embodiment. For example, the end point t2 may be a point at which a predetermined time has elapsed after the detected value of each portion becomes substantially constant.

The definitions of the heel section Ah, the intermediate section Am, and the toe section At are not limited to the above embodiments. For example, the section from the time when the detected value of the toe sensor 12b becomes the apex of the peak to the end point t2 of the step section A may be set as the toe section At. Further, the section from the start point t1 of the step section A to the end point t2 is set as the heel section Ah, and the section from before the specific time of the end point t2 of the step section A to the end point t2 is set as the toe section At. Each subsection may be set without using the relative pressure relationship. In these cases, the toe section At can be reliably provided.

-The method of dividing the one-step section A into a plurality of subsections is not limited to the above embodiment. For example, the intermediate section Am may be further divided into 4 or more small sections. It may be a small section divided independently of the section where pressure is applied to the heel side and the section where pressure is applied to the toe side.

-The method of dividing the two-dimensional coordinate system at the center of foot pressure when calculating the stay ratio is not limited to the above embodiment. For example, in the above embodiment, the region is divided into a plurality of regions arranged in the left-right direction, but the region may be divided into a plurality of regions arranged in the front-rear direction, or may be divided into a plurality of regions arranged in the front-rear and left-right directions. Further, the number of divisions of the two-dimensional coordinate system at the center of foot pressure may be 2 or 4 or more.

-In the above embodiment, the runout values of all the subsections have been acquired, but the configuration may be such that only the runout values of a specific subsection are acquired.
-The type of the feature amount that constitutes the feature amount data for performing clustering is not limited to the above embodiment, and may include a runout value in units of small sections. For example, the feature amount data may consist only of the runout values Sh, Sm, St, the heel section maximum value Ah _max , the toe section maximum value At _max , and the section time Th, Tm, Tt, or the feature related to a specific small section. It may be feature data consisting only of quantities. Further, in addition to the feature amounts described in the above-described embodiment, the feature amount data may include other feature amounts. Examples of the other feature amount include FTA, LHA, and the maximum value of the detected values of the four pressure sensors 12 in the intermediate section Am.

-The information processing unit 23 may acquire a feature amount that is not used for the determination of the determination unit. The feature amount that is not used for the judgment of the judgment unit also helps the medical staff to appropriately grasp the symptom of the person to be measured when examining the person to be measured.

Next, the technical idea that can be grasped from the above-described embodiment and modification is described below.
(B) A walking diagnostic device that diagnoses the walking state, and information processing that acquires the feature amount related to the walking state from the measurement information obtained by measuring the pressure of each part applied to the sole of one foot over time. A unit, a determination unit that determines whether or not the walking state is a predetermined alert target state based on the acquired feature amount, and a determination unit that determines whether the walking state is a predetermined alert target, and a walking state is a predetermined alert target based on the determination result of the determination unit. A gait diagnostic device including a notification unit for notifying the state of the above.

(B) A program that causes a computer to function as a gait diagnostic device, and is a step of acquiring features related to the gait state from measurement information including measurement values obtained by measuring the pressure of each part on the sole of one foot over time. The step of determining whether or not the walking state is the state of the predetermined alert target based on the acquired feature amount, and the state of the walking state of the predetermined alert target based on the determination result of the determination unit. A program characterized by having a computer perform a step of notifying that it is.

10 ... Measuring device 20 ... Walking diagnostic device 23 ... Information processing unit 24 ... Judgment processing unit 25 ... Display unit 32 ... Dividing unit 35 ... Coordinate calculation unit 37 ... Runout value calculation unit 39 ... FTA calculation unit 40 ... LHA calculation unit 51 ... 1st judgment unit 52 ... 2nd judgment unit 53 ... 3rd judgment unit

Claims (5)

It is a walking diagnosis system that diagnoses walking conditions.
A measuring unit that measures the pressure of each part on the sole of one foot over time,
An information processing unit that acquires features related to the walking state from the measurement information measured by the measurement unit, and
A determination unit that determines whether or not the walking state is a predetermined alert target state based on the acquired feature amount, and
A walking diagnosis system including a notification unit that notifies that the walking state is a predetermined alert target state based on the determination result of the determination unit. The determination unit determines whether or not the walking state is a predetermined alert target state based on the comparison between the acquired feature amount and the alert information preset for each person to be measured by the medical staff. The walking diagnosis system according to claim 1. The information processing unit includes an FTA calculation unit that acquires a femoral tibial angle as the feature amount.
The walking diagnosis system according to claim 1 or 2, wherein the determination unit determines whether or not the walking state is a predetermined alert target state based on the acquired femoral tibial angle. The information processing unit includes an LHA calculation unit that acquires a leg heel angle as the feature amount.
The walking diagnosis system according to any one of claims 1 to 3, wherein the determination unit determines whether or not the walking state is a predetermined alert target state based on the acquired leg heel angle. The information processing unit
A division that divides the one-step section from the landing of the foot to the takeoff during walking into multiple subsections,
A coordinate calculation unit that calculates the two-dimensional coordinates of the foot pressure center of the sole of the foot at regular intervals in the small section, and
In the same within the sub-interval, in turn coordinate P _n three of the two-dimensional coordinates of consecutive predetermined time _intervals, the coordinate P _{n + 1,} when the coordinates P _{n + 2,} show a variation of the coordinate P _{n + 1} from coordinate P _n It is provided with a runout value calculation unit that calculates a runout value, which is the sum of the runout angles formed by the vector and the vector indicating the amount of change from the coordinates P _{n + 1} to the coordinates P _{n + 2, as the feature amount.}
The walking diagnosis system according to any one of claims 1 to 4, wherein the determination unit determines whether or not the walking state is a predetermined alert target state based on the acquired runout value.

Images