Perception support system, evaluation device, and program therefor

Abstract

To predict the future risk of knee deformation caused by gonarthrosis due to acquired factors, and contribute to the prevention of the knee deformation.SOLUTION: A perception support system according to the present invention comprises: a detection device 12 that detects the plantar pressure of a subject; an evaluation device 13 that performs an evaluation regarding the risk of knee deformation of the subject on the basis of the plantar pressure data; and a presentation device 14 that presents information and stimuli related to the evaluation results of the evaluation device 13. The evaluation device 13 includes: a COP calculation unit 19 that obtains COP representing the central position of the plantar pressure during the subject's walking motion; a COP inclination angle calculation unit 20 that determines a COP straight line connecting respective points of COP at two different times within one stance phase and determines a COP inclination angle being an angle of the COP straight line with respect to the longitudinal axis of the sole; and an information creation unit 21 that creates information related to the deformation risk on the basis of the COP inclination angle.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a perception support system, an evaluation device, and a program therefor that predict the risk of knee deformity of a subject and contribute to preventing the knee deformity. [Background technology]

[0002] There are various causes of knee pain associated with aging, including osteoarthritis of the knee. Osteoarthritis of the knee is a condition characterized by pain and swelling caused by direct friction of the knee joint due to wear and tear of the cartilage tissue in the knee joint. In the early stages of osteoarthritis, pain is felt upon initiating movement. Progression leads to constant pain and joint effusion, which can interfere with walking. Further progression leads to deformities such as flexion contracture. In the final stages, walking ability declines significantly, leading to a decline in activities of daily living (ADL) and quality of life (QOL). For these reasons, early treatment of osteoarthritis of the knee is important for extending healthy lifespans and reducing medical costs. Osteoarthritis of the knee can be classified into three types: medial type (bow-legged), which primarily affects the inside of the knee; lateral type (knock-legged), which primarily affects the outside of the knee; and mixed type, which affects both the inside and outside of the knee.

[0003] Incidentally, Patent Document 1 discloses a method for determining the risk of knee osteoarthritis, and Patent Document 2 discloses a walking assistance device for patients with knee osteoarthritis. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2020-178585 [Patent Document 2] Japanese Patent Publication No. 2023-45150 Summary of the Invention [Problem to be solved by the invention]

[0005] The method of Patent Document 1 focuses on genetic factors that affect physical constitution and predicts the risk of developing knee osteoarthritis based on congenital genetic factors using genetic analysis. Therefore, it cannot predict the risk of developing knee osteoarthritis based on acquired factors such as the application of load to the lower limbs during walking, which is strongly related to the development of knee osteoarthritis. Furthermore, the device of Patent Document 2 is an orthosis that limits the range of motion of the knee when the load on the lower limbs increases, thereby limiting the load on the knee. However, it does not predict the risk of developing knee osteoarthritis and encourages gait modifications based on the risk.

[0006] According to the research of the present inventors, the angle between the line connecting two points on the trajectory of the center of plantar pressure during one walking cycle and the long axis of the plantar is one factor that affects the risk of knee deformity leading to osteoarthritis.

[0007] The present invention was devised based on the above-mentioned problems and findings, and its purpose is to provide a perception support system, evaluation device, and program therefor that can predict the risk of future knee deformity due to osteoarthritis caused by acquired factors and contribute to the prevention of such knee deformity. [Means for solving the problem]

[0008] In order to achieve the above-mentioned object, the present invention mainly comprises a detection device that detects the plantar pressure of a subject, and an evaluation device that evaluates the risk of knee deformity of the subject based on the plantar pressure data, and the evaluation device evaluates the risk of deformity based on the COP inclination angle, which is the angle between the long axis of the plantar and a COP line connecting two different points on a COP trajectory that represents the time-series displacement of the center position of the plantar pressure during the walking movement of the subject. [Effects of the Invention]

[0009] According to the present invention, by detecting the plantar pressure at the time of contact with the ground during walking of a subject, it is possible to predict the future risk of knee deformity, taking into account acquired factors such as gait, from the degree of outward inclination of the foot of the COP trajectory, which is the trajectory of the center position of the plantar pressure during stance. Furthermore, the presentation device can make the subject aware of the deformity risk, and various information regarding correction of the subject's movements, etc., to prevent future knee deformity can be presented to the subject, physical therapists, etc. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a block diagram showing a schematic configuration of a perception support system according to an embodiment of the present invention. [Figure 2] FIG. 2 is a conceptual diagram for explaining the arrangement of pressure sensors. [Figure 3] FIG. 1 is a conceptual diagram of the sole of the foot for explaining the COP inclination angle. [Figure 4] FIG. 10 is a conceptual diagram of a deformation risk prediction map. [Figure 5] FIG. 10 is a conceptual diagram of a sole portion for explaining a modified example regarding calculation of the COP tilt angle. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0012] A block diagram showing the schematic configuration of a perception support system according to this embodiment is shown in Figure 1. In this figure, the perception support system 10 is a system that supports a subject's perception of knee deformity that leads to medial osteoarthritis, commonly known as bowlegs, and is a system that predicts the risk of knee deformity leading to bowlegs and contributes to the prevention of bowlegs based on the risk of deformity.

[0013] The perception support system 10 includes a detection device 12 that detects the plantar pressure of the subject during walking, an evaluation device 13 that evaluates the subject's risk of knee deformity based on the plantar pressure data, and a presentation device 14 that presents information and stimuli related to the evaluation results of the evaluation device 13.

[0014] As shown in Fig. 2, the detection device 12 comprises pressure sensors 17 provided at multiple locations on an insole 16 with which the sole of the subject comes into contact. Although not particularly limited, multiple pressure sensors 17 are provided in each of three areas A1 to A3 to which load is mainly applied during the stance phase. In this embodiment, two pressure sensors 17 are provided in area A1 near the big toe and area A2 near the lateral metatarsophalangeal joint (MP joint), and four pressure sensors 17 are provided in area A3 near the heel. Each pressure sensor 17 transmits an electrical signal corresponding to the applied load to the evaluation device 13 sequentially via wire or wirelessly.

[0015] As will be described later, as long as it is possible to detect the center position of the plantar pressure during the subject's walking movement, the installation position and number of pressure sensors 17 are not limited to the embodiment described above, and various embodiments can be adopted, and various sensors, devices, etc. can also be used as the detection device 12.

[0016] The evaluation device 13 is configured by a computer including a processing unit such as a CPU, and storage devices such as a memory and a hard disk, and has installed thereon programs for causing the computer to function as the following units.

[0017] As shown in FIG. 1, this evaluation device 13 includes a COP calculation unit 19 that calculates the center of pressure (COP) of the plantar pressure during the stance phase of the subject's walking movement from the detection results of each pressure sensor 17, a COP tilt angle calculation unit 20 that calculates the COP tilt angle, which is the angle corresponding to the COP trajectory that represents the time-series displacement of the COP during one gait cycle (one stance phase), and an information creation unit 21 that creates information related to the risk of bow-leg deformation based on the COP tilt angle.

[0018] The COP calculation unit 19 calculates the coordinates (X ) of the COP, which is the center position of the sole pressure of the subject, from the load detected by each pressure sensor 17 using the following equation: G , Y G ) are calculated sequentially. In this coordinate system, as shown in FIG. 2, the coordinate axis in the left-right direction (horizontal direction) of the sole is the X axis, and the coordinate axis in the up-down direction (vertical direction) of the sole is the Y axis.

number

[0019] As described above, during one stance phase of one gait cycle of one foot of the subject, the loads at eight locations on the sole of the foot are measured at predetermined time intervals by each pressure sensor 17, and the COP is calculated for each measurement value acquisition time. As a result, the COP trajectory C in Fig. 3, which extends from the landing point P1 near the heel to the take-off point P2 near the toe, is identified as the trajectory representing the time-series change of the COP.

[0020] The COP tilt angle calculation unit 20 calculates a COP line S connecting two COP points at any two different times on the COP trajectory C, and then calculates a COP tilt angle θ, which is the angle of the COP line S with respect to the long axis L of the sole. In this embodiment, one point for determining the COP line S is the first landing point P1 on the COP trajectory C, and the other point is the COP acquired a predetermined time T after the landing point P1. This creates a COP line S near the landing point P1 where there is little disturbance, such as the influence of hallux valgus, and enables more accurate assessment of the risk of deformation, which will be described later.

[0021] The information creation unit 21 includes a memory unit 23 that stores data on the risk of bow-leg deformation corresponding to the COP tilt angle θ, a prediction unit 24 that predicts the future risk of bow-leg deformation from the COP tilt angle θ identified in the current walking movement of the subject, and a prevention support unit 25 that issues a presentation command to the presentation device 14 to prevent bow-leg deformation from the current COP tilt angle θ.

[0022] The storage unit 23 stores a deformation risk prediction map that has been created in advance by accumulating and learning time-series data on the COP tilt angle θ of various subjects. As shown in FIG. 4, this deformation risk prediction map is a database that associates age, which is time information, with the COP tilt angle θ and the risk of bow-leg deformation. The database is constructed by machine learning or the like, using data for each subject that correlates the subject's age with the COP tilt angle θ and the actual bow-leg condition at the time of walking for many subjects who have previously walked. Note that the time series here can be based on various perspectives, such as the subject's age, the time elapsed since the start of hospital visits, etc.

[0023] In the deformity risk prediction map in Figure 4, areas on the map where the probability of bow legs, that is, the medial malleolus of the femur (knee) not touching when the medial malleolus of the ankle joint (ankle) is aligned, is 80% or higher, and these areas are considered to be at "high deformity risk." Areas on the map where this probability is 20% or lower are considered to be at "low deformity risk," and areas between these are considered to be at "medium deformity risk."

[0024] The prediction unit 24 predicts future changes in deformity risk over time for a subject whose bow-legged deformity risk is predicted by inputting the COP tilt angle θ acquired over a predetermined period of time using a prediction model such as an LSTM (Long Short Term Memory) model that uses a deformity risk prediction map. For example, in the deformity risk prediction map shown in FIG. 4, the solid line represents the subject's (patient's) input value, and the dashed line represents the calculated future predicted value. According to this example, Patient A is currently at "medium risk of deformity," but if no preventive measures are taken and the current gait state is maintained, it is predicted that the patient will transition to "high risk of deformity" as the patient ages. In such a case, it is determined that some kind of action is urgently needed. On the other hand, Patient B's level of "medium risk of deformity" has declined, and even if the patient continues his current gait state, it is predicted that the patient will transition to "low risk of deformity" as the patient ages, and it is therefore determined that the current situation is not problematic.

[0025] The prediction unit 24 may also use other machine learning models, statistical models, etc. to predict the risk of deformity. Furthermore, a prediction model that adds prior data acquisition of severity indices of knee osteoarthritis to the above model may output a time-dependent prediction of the severity index of future knee deformity in addition to the deformity risk. Examples of severity indices include the Kellgren-Lawrence classification (KL classification), the medial proximal tibial angle (MPTA), and the % mechanical axis (%MA).

[0026] The prediction unit 24 predicts the future risk of bow-legged deformity for the subject as a healthy individual. It can also predict the risk of deformity for postoperative evaluation of knee osteotomy, which surgically corrects the skeletal shape of a bow-legged patient to a normal state. Patients who have undergone knee osteotomy are more likely to develop bow-legged deformity than healthy individuals for a certain period after surgery due to fragile tissues. Therefore, when a patient who has undergone knee osteotomy is the subject, the prediction unit 24 predicts the risk of bow-legged deformity as the risk of postoperative recurrence as follows: In this case, in addition to age and COP tilt angle θ, input values ​​include the number of days since surgery, the number of days since the start of rehabilitation, and the surgical correction angle, etc., to create a deformity risk prediction map. The prediction model then calculates the risk of recurrence over time based on the patient's input values ​​up to the present time, and can also predict various severity indicators.

[0027] The prevention support unit 25 performs all or part of the following processing to guide the subject to proper walking based on the COP tilt angle θ calculated by the COP tilt angle calculation unit 20.

[0028] In the first process, the COP tilt angle θ for each step is compared with a preset threshold value for the outward movement of the foot, and it is determined whether or not the threshold value is exceeded. Then, depending on the degree of the exceedance, a command is sent to the presentation device 14, and information or stimuli are presented to the subject or the like via the presentation device 14.

[0029] In the second process, data on the COP tilt angle θ during the subject's rehabilitation, etc. is accumulated and saved, and divided into predetermined time periods, such as short-term periods in units of the number of steps taken during rehabilitation, or medium- to long-term periods in units of years and months. For each period, the number of times the COP tilt angle θ exceeds the threshold value in the first process, the average COP tilt angle θ, etc. are derived, and this period data is presented as feedback to the subject, physical therapist, etc. via presentation device 14.

[0030] In the third process, information and stimuli are presented to the subject or the like through the presentation device 14 in accordance with the result of the prediction by the prediction unit 24 of the risk of future bow-leg deformation.

[0031] The presentation device 14 comprises a transmission device capable of transmitting information and stimuli according to the processing results of the prevention support unit 25 to the subject, etc., so as to guide the subject, etc., to walk properly via sound, images, and vibration. This transmission device may be any device or system capable of transmitting sound, images, vibration, etc. to the subject, etc., and examples thereof include a speaker that outputs sound audible to the subject, a personal computer or smartphone display that displays images visible to the subject, and a vibration belt that is worn on the arm or waist of the subject, etc., and applies vibrations to the part where the device is worn that the subject, etc. can perceive.

[0032] In the presentation device 14, when the COP tilt angle θ for each step exceeds a threshold, the first process in the prevention support unit 25 provides biofeedback according to the degree of exceedance. Examples of feedback include changing the pitch of a sound, the color on the screen, the magnitude of vibration, and the like. For example, when the COP tilt angle θ significantly exceeds the threshold, the presentation device 14 is operated to provide high-intensity feedback to make the subject aware of the large inward load in the next step. When the COP tilt angle θ slightly exceeds the threshold, the presentation device 14 is operated to provide low-intensity feedback to make the subject aware of the slight inward load in the next step. When the COP tilt angle θ does not exceed the threshold, no feedback is provided, and the subject is informed that their current walking is correct. It is also possible to further strengthen the feedback when the COP tilt angle θ repeatedly exceeds the threshold.

[0033] Furthermore, in the case of the second process in the prevention support unit 25, the aforementioned period data is sequentially transmitted to the subject or physical therapist via voice, image, etc., allowing the physical therapist to provide the subject with verbal feedback or visual feedback via images such as graphs and tables, etc. This is useful for improving gait during rehabilitation and for creating and improving a mid- to long-term rehabilitation plan spanning several months or several years.

[0034] Furthermore, in the case of the third processing by the prevention support unit 25, suggestions regarding the need for improvements to prevent bow-leg deformation are presented depending on the degree of risk of bow-leg deformation in the future as estimated by the prediction unit 24, and the presentation intensity is adjusted by the presentation device 14. In the example of Fig. 4, since patient A is predicted to have a high risk of bow-leg deformation in the future, a preventive system for gait improvement such as the first processing by the prevention support unit 25 is required to prevent bow-leg deformation from the present, and the presentation intensity by the presentation device 14 is set to increase the system's intensity. On the other hand, since patient B is predicted to have a low risk of bow-leg deformation in the future, the presentation device 14 presents a message to the subject or the like that no preventive measures for bow-leg deformation are currently required.

[0035] As shown in Fig. 5, the COP tilt angle calculation unit 20 can also extract multiple COP lines S with different combinations of the acquisition times of two points on the COP trajectory C to determine multiple COP tilt angles θ. Here, because disturbance factors such as hallux valgus are greater closer to the toes, it is better to extract more points near the landing point P1, which purely reflects gait characteristics. In this way, by comparing the predicted results of bow-leg deformation risk with the actual situation for multiple COP line S patterns, it is possible to select points that have a high degree of influence on the deformation risk, and to identify the elapsed time from the landing point P1 when selecting another point other than the landing point P1 when determining the COP line S.

[0036] Furthermore, when multiple COP tilt angles θ are calculated by the COP tilt angle calculation unit 20, adding these to the learning data and original data for prediction of the prediction model allows for more complex gait characteristics to be reflected, making it possible to create a more accurate prediction model. In this case, the dominance of each COP tilt angle θ on the deformity risk can be calculated using statistics, deep learning, etc., and a weighting can be set according to this dominance, making it possible to create a more accurate prediction model for bow-leg deformity that integrates the COP tilt angles θ on each COP line S. Here, a simple model that can be operated by measuring only the areas with a high dominance can be created, taking into account the processing power, measurement equipment, environment, etc. of the PC where bow-leg deformity is assessed.

[0037] Furthermore, the perception support system 10 of the present invention can be used to predict and prevent future risks of knee deformation in general knee osteoarthritis, and can also be applied as a system to support the subject's perception in general autonomous movement, not only in the subject's walking but also in predicting and preventing knee deformity disorders in sports that involve running.

[0038] Furthermore, the configuration of each part of the device in the present invention is not limited to the illustrated configuration example, and various modifications are possible as long as they provide substantially the same effect. [Explanation of symbols]

[0039] 10 Perception Support Systems 12 Detection device 13 Evaluation equipment 14 Presentation device 19 COP calculation section 20 COP tilt angle calculation section 21 Information Creation Department 23 Memory section 24 Prediction Department 25 Prevention Support Department C COP trajectory P1 landing point L long axis S COP straight line θ COP inclination angle

Claims

1. a detection device for detecting a plantar pressure of a subject; and an evaluation device for evaluating a risk of knee deformity of the subject based on the plantar pressure data; The evaluation device performs an evaluation of the risk of deformation based on the COP tilt angle, which is the angle between the long axis of the sole and a COP line connecting two different points on a COP trajectory that represents the time-series displacement of the center position of the plantar pressure during the walking movement of the subject. This is a perception support system characterized by the above.

2. a detection device for detecting a plantar pressure of a subject; and an evaluation device for evaluating a risk of knee deformity of the subject based on the plantar pressure data; The evaluation device is a perception support system characterized by comprising: a COP calculation unit that calculates the COP, which is the center position of the plantar pressure during the walking movement of the subject; a COP tilt angle calculation unit that calculates a COP line connecting each point of the COP at two different times during one stance phase and then calculates the COP tilt angle, which is the angle of the COP line with respect to the long axis of the plantar; and an information creation unit that creates information regarding the deformation risk based on the COP tilt angle.

3. 3. The perception support system according to claim 2, wherein the information generating unit includes a predicting unit that predicts the future risk of deformation from the COP tilt angle during walking of the subject.

4. The information creation unit further includes a storage unit in which a deformation risk prediction map in which time information, the COP tilt angle, and the deformation risk are associated with each other is stored, The deformation risk prediction map is created in advance by accumulating and learning time-series data of the COP tilt angles of various subjects, 4. The perception support system according to claim 3, wherein the prediction unit predicts the deformation risk from the input COP tilt angle using the deformation risk prediction map.

5. 5. The perception support system according to claim 3, wherein the prediction unit also predicts a severity index regarding future knee deformity of the subject.

6. The perception support system according to claim 2, wherein the COP tilt angle calculation unit calculates the COP straight line by connecting a point on the COP near the subject's landing point with a point a predetermined time after the landing point.

7. The perception support system according to claim 2, characterized in that the COP tilt angle calculation unit extracts a plurality of the COP straight lines based on different combinations of two points on a COP locus that represents a time-series change of the COP, and calculates a plurality of the COP tilt angles.

8. The evaluation device further includes a presentation device that presents information and / or stimuli related to the evaluation result by the evaluation device, 4. The perception support system according to claim 3, wherein the information creation unit includes a prevention support unit that issues a presentation command to the presentation device for preventing deformation of the knee based on the COP tilt angle.

9. 9. The perception support system according to claim 8, wherein the presentation device presents the information and / or the stimulus to the subject so as to guide the subject to walk in a proper manner.

10. The perception support system according to claim 8, characterized in that the prevention support unit issues the presentation command to adjust the intensity of the information for preventing deformation and / or the stimulation depending on the prediction result of the deformation risk by the prediction unit.

11. An apparatus for evaluating a subject's risk of knee deformity based on data on the subject's plantar pressure, comprising: An evaluation device characterized by comprising: a COP calculation unit that calculates the COP, which is the center position of the plantar pressure during the walking movement of the subject; a COP tilt angle calculation unit that calculates a COP line connecting each point of the COP at two different times during one stance phase and then calculates the COP tilt angle, which is the angle of the COP line with respect to the long axis of the plantar; and an information creation unit that creates information regarding the deformation risk based on the COP tilt angle.

12. A program for an apparatus for evaluating a subject's risk of knee deformity based on the subject's plantar pressure data, A program for an evaluation device that causes a computer to function as a COP calculation unit that calculates the COP, which is the center position of the plantar pressure during the walking movement of the subject, a COP tilt angle calculation unit that calculates a COP line connecting each point of the COP at two different times during one stance phase and then calculates the COP tilt angle, which is the angle of the COP line with respect to the long axis of the plantar, and an information creation unit that creates information regarding the deformity risk based on the COP tilt angle.

Images