Index value estimation device, estimation system, index value estimation method, and program

Abstract

To provide an index value estimation device, etc. capable of appropriately estimating an index value indicating a state of the knee in a daily life.SOLUTION: An index value estimation device includes: a data acquisition unit for acquiring feature amount data including a feature amount used for estimating an index value indicating a state of the knee of a user, which is extracted from sensor data on a motion of the user's foot; a storage unit for storing an estimation model for outputting an index value according to input of the feature amount data; an estimation unit for estimating output acquired by inputting the acquired feature amount data to the estimation model as an index value indicating the state of the user's knee; and an output unit for outputting information on the estimated index value indicating the state of the user's knee.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to an index value estimation device that estimates an index value indicating the state of the knee, etc.

Background Art

[0002] With the increasing interest in healthcare, attention has been focused on services that provide information according to walking gait. For example, technologies for analyzing walking gait using sensor data measured by sensors mounted on footwear such as shoes have been developed. In the time-series data of the sensor data, features associated with walking events related to the body state appear. By analyzing the walking data including the features associated with the walking events, the body state of the subject can be estimated. For example, if the state of the subject's knee can be estimated, early detection and prevention of diseases such as osteoarthritis become possible.

[0003] Patent Document 1 discloses a knee state determination system that determines the knee state of a user by focusing on the movement of the knee part during footstep. The system of Patent Document 1 includes a plurality of sensor devices and a knee state estimation device. The plurality of sensor devices are each mounted on the waist, the thighs of both legs, and the lower legs of both legs. The plurality of sensor devices measure the angular velocity generated by the rotational movement of the thighs and lower legs accompanying the user's footstep. The plurality of sensor devices transmit the rotational angular velocity reflecting the measured angular velocity to the knee state determination device. The knee state determination device determines the knee state of the user by analyzing the data transmitted from the sensor devices. Specifically, the knee state determination device performs an abnormality determination of the user's knee using the yaw direction component around the axis in the direction of gravity output from each sensor device mounted on the thighs and lower legs of both legs.

[0004] Patent Document 2 discloses a detection device used to estimate the state of a person during movement. The device in Patent Document 2 includes sensors such as an acceleration sensor and an angular velocity sensor. The device in Patent Document 2 is attached to the knee or around the knee of a subject. The acceleration detected by the device in Patent Document 2 is used to estimate the state of the knee. Patent Document 2 discloses that the detected acceleration can be used to estimate the degree and prognosis of osteoarthritis of the knee.

[0005] Non-patent documents 1-4 contain various reports on the verification of knee conditions related to diseases such as osteoarthritis of the knee. Non-patent document 1 reports on the development of diagnostic methods for knee diseases such as osteoarthritis of the knee. Non-patent document 1 lists height, leg length, range of motion, and lower limb alignment as factors that influence gait patterns. Non-patent document 2 reports on the results of a gait analysis conducted on multiple subjects for the purpose of quantitatively evaluating lateral thrust observed in patients with osteoarthritis of the knee. Non-patent document 3 reports on the results of evaluating gait abnormalities caused by osteoarthritis of the knee in multiple subjects. Non-patent document 4 reports on the results of an investigation into the muscles that affect knee flexion velocity during the bilateral support phase of walking. Non-patent document 4 states that if the knee flexion velocity is sufficient at toe-off, appropriate knee flexion can be obtained in the swing phase. [Prior art documents] [Patent Documents]

[0006] [Patent Document 1] Japanese Patent Publication No. 2016-106948 [Patent Document 2] Japanese Patent Publication No. 2022-051451 [Non-patent literature]

[0007] [Non-Patent Document 1] Yuki Ishikawa et al., “A Proposal for the Construction of a Knee Disease Diagnostic Method Using Individual Modeling - Towards Elucidating the Pathogenesis of Osteoarthritis of the Knee -”, Proceedings of the 2012 JSME Conference on Robotics and Mechatronics, Hamamatsu, Japan, May 27-29, (2012), pp. 2P1-I02(1)-2P1-I02(2). [Non-Patent Document 2] Takashi Komura et al., "A Study on Gait Analysis in Patients with Medial-Type Osteoarthritis of the Knee," Journal of the Faculty of Medicine, Kobe University, 61(4), (2001), pp. 89-94. [Non-Patent Document 3] Shunsuke Yamashina, "Development of an observational method for evaluating gait abnormalities in patients with osteoarthritis of the knee undergoing conservative treatment and verification of its relationship with reduced physical activity," Doctoral dissertation, Kibi International University, 2019. [Non-Patent Document 4] SR Goldberg, et al., Journal of Biomechanics, 37, (2004), pp.1189-1196. [Overview of the Initiative] [Problems that the invention aims to solve]

[0008] The method described in Patent Document 1 estimates the user's knee condition using angular velocity measured by multiple sensors attached to the body. However, the method in Patent Document 1 requires sensors to be attached to multiple locations on the waist and legs. Therefore, the method in Patent Document 1 is not easily applicable to estimating the user's knee condition in daily life.

[0009] Patent Document 2 discloses a method for estimating the condition of the knee using acceleration and angular velocity measured by an acceleration sensor attached to or around the knee of a subject. In the method of Patent Document 2, the sensor is attached to or around the knee using a flexible assistive device. Therefore, the method of Patent Document 2 is not suitable for applications that require timely estimation of the user's knee condition because the sensor attachment position tends to change from day to day.

[0010] As described in Non-Patent Documents 1-4, large-scale equipment can be used to examine diseases such as osteoarthritis of the knee in detail. However, methods like those described in Non-Patent Documents 1-4 require large-scale equipment, making them difficult to apply to estimating knee condition in daily life.

[0011] The purpose of this disclosure is to provide an index value estimation device, etc., that can appropriately estimate index values ​​indicating the condition of the knee in daily life. [Means for solving the problem]

[0012] An index value estimation device according to one aspect of the present disclosure includes: a data acquisition unit that acquires feature data including features used to estimate an index value indicating the state of the user's knee, extracted from sensor data relating to the movement of the user's feet; a storage unit that stores an estimation model that outputs an index value corresponding to the input of feature data; an estimation unit that estimates the output obtained by inputting the acquired feature data into the estimation model as an index value indicating the state of the user's knee; and an output unit that outputs information regarding the estimated index value indicating the state of the user's knee.

[0013] In one aspect of the present disclosure, a method for estimating an index value involves acquiring feature data, which includes features extracted from sensor data relating to the user's foot movements and used to estimate an index value indicating the user's knee condition; inputting the acquired feature data into an estimation model that outputs an index value corresponding to the input of the feature data; estimating the output obtained by inputting into the estimation model as an index value indicating the user's knee condition; and outputting information regarding the estimated index value indicating the user's knee condition.

[0014] A program according to one aspect of the present disclosure causes a computer to execute a process of obtaining feature amount data including feature amounts used for estimating an index value indicating a state of a user's knee, which is extracted from sensor data related to the movement of the user's feet, a process of inputting the obtained feature amount data into an estimation model that outputs an index value corresponding to the input of the feature amount data, a process of estimating, as an index value indicating the state of the user's knee, an output obtained by inputting the feature amount data into the estimation model, and a process of outputting information related to the estimated index value indicating the state of the user's knee.

Advantages of the Invention

[0015] According to the present disclosure, it becomes possible to provide an index value estimation device or the like that can appropriately estimate an index value indicating the state of a knee in daily life.

Brief Description of the Drawings

[0016] [Figure 1] It is a block diagram showing an example of the configuration of an estimation system according to the first embodiment. [Figure 2] It is a block diagram showing an example of the configuration of a measurement device included in the estimation system according to the first embodiment. [Figure 3] It is a conceptual diagram showing an example of the arrangement of the measurement device according to the first embodiment. [Figure 4] It is a conceptual diagram for explaining an example of the relationship between the local coordinate system and the world coordinate system set in the measurement device according to the first embodiment. [Figure 5] It is a conceptual diagram for explaining the human body surface used in the explanation of the measurement device according to the first embodiment. [Figure 6] It is a conceptual diagram for explaining the walking cycle used in the explanation of the measurement device according to the first embodiment. [Figure 7] It is a graph for explaining an example of the time-series data of the sensor data measured by the measurement device according to the first embodiment. [Figure 8]It is a diagram for explaining an example of normalization of walking waveform data extracted from time-series data of sensor data measured by a measurement device according to the first embodiment. [Figure 9] It is a block diagram showing an example of the configuration of an index value estimation device included in an estimation system according to the first embodiment. [Figure 10] It is a graph for explaining parameters related to the knee joint flexion angle estimated by an estimation system according to the first embodiment. [Figure 11] It is a table summarizing an example of feature amounts used for estimation of parameters related to the knee joint flexion angle estimated by an estimation system according to the first embodiment. [Figure 12] It is a table summarizing an example of feature amounts used for estimation of parameters related to the knee joint flexion angle estimated by an estimation system according to the first embodiment. [Figure 13] It is a table summarizing an example of feature amounts used for estimation of parameters related to the knee joint flexion angle estimated by an estimation system according to the first embodiment. [Figure 14] It is a table summarizing an example of feature amounts used for estimation of parameters related to the knee joint flexion angle estimated by an estimation system according to the first embodiment. [[ID=一十九]] [Figure 15] It is a table summarizing an example of feature amounts used for estimation of parameters related to the knee joint flexion angle estimated by an estimation system according to the first embodiment. [Figure 16] It is a flowchart for explaining an example of the operation of a measurement device included in an estimation system according to the first embodiment. [Figure 17] It is a flowchart for explaining an example of the operation of an index value estimation device included in an estimation system according to the first embodiment. [Figure 18] It is a conceptual diagram for explaining an application example of an estimation system according to the first embodiment. [Figure 19] It is a block diagram showing an example of the configuration of an estimation system according to the second embodiment. [Figure 20]This graph illustrates the Angular Jerk Cost estimated by the estimation system according to the second embodiment. [Figure 21] This is a flowchart illustrating an example of the operation of a measuring device included in the estimation system according to the second embodiment. [Figure 22] This is a flowchart illustrating an example of the operation of the index value estimation device included in the estimation system according to the second embodiment. [Figure 23] This is a conceptual diagram illustrating an example of the application of the estimation system according to the second embodiment. [Figure 24] This is a block diagram showing an example of the configuration of an index value estimation device according to the third embodiment. [Figure 25] This block diagram shows an example of a hardware configuration that performs control and processing for each embodiment. [Modes for carrying out the invention]

[0017] The embodiments for carrying out the present invention will be described below with reference to the drawings. However, the embodiments described below have technically preferred limitations for carrying out the present invention, but the scope of the invention is not limited thereto. In all the figures used in the description of the embodiments below, the same parts are denoted by the same reference numerals unless there is a particular reason not to. Also, in the embodiments below, repeated explanations of similar configurations and operations may be omitted.

[0018] (First Embodiment) First, the estimation system according to the first embodiment will be described with reference to the drawings. The estimation system of this embodiment measures sensor data related to the movement of the user's feet in accordance with their walking. The estimation system of this embodiment uses the measured sensor data to estimate an index value indicating the state of the user's knee. In this embodiment, an example is given in which a parameter related to the knee joint flexion angle is estimated as an index value indicating the state of the knee. The knee joint flexion angle is the angle formed by the thigh and the lower leg with the knee joint as the center. In this embodiment, the knee joint flexion angle is the angle in the plane of the direction of movement (sagittal plane).

[0019] (composition) Figure 1 is a block diagram showing an example of the configuration of the estimation system 1 according to this embodiment. The estimation system 1 comprises a measuring device 10 and an index value estimation device 13. In this embodiment, an example in which the measuring device 10 and the index value estimation device 13 are configured on separate hardware will be described. For example, the measuring device 10 is installed on the footwear of the subject (user) whose index value indicating the condition of the knee is the target of estimation. For example, the functions of the index value estimation device 13 are installed on a portable terminal carried by the subject (user). In the following, the configurations of the measuring device 10 and the index value estimation device 13 will be described individually.

[0020] [Measuring device] Figure 2 is a block diagram showing an example of the configuration of the measuring device 10. The measuring device 10 includes a sensor 11 and a feature data generation unit 12. In this embodiment, an example is given in which the sensor 11 and the feature data generation unit 12 are integrated. The sensor 11 and the feature data generation unit 12 may be provided as separate devices. For example, the feature data generation unit 12 may be incorporated into the index value estimation device 13. In that case, the measuring device 10 transmits the sensor data measured by the sensor 11 to the index value estimation device 13.

[0021] As shown in Figure 2, sensor 11 includes an acceleration sensor 111 and an angular velocity sensor 112. Figure 2 shows an example in which the acceleration sensor 111 and the angular velocity sensor 112 are included in sensor 11. Sensor 11 may also include sensors other than the acceleration sensor 111 and the angular velocity sensor 112. The description of sensors other than the acceleration sensor 111 and the angular velocity sensor 112 that may be included in sensor 11 is omitted.

[0022] The acceleration sensor 111 is a sensor that measures acceleration in three axes (also called spatial acceleration). The acceleration sensor 111 measures acceleration (also called spatial acceleration) as a physical quantity related to the movement of the foot. The acceleration sensor 111 outputs the measured acceleration to the feature data generation unit 12. For example, the acceleration sensor 111 can be a piezoelectric, piezoresistive, or capacitive type sensor. The sensor used as the acceleration sensor 111 is not limited to any measurement method as long as it can measure acceleration.

[0023] The angular velocity sensor 112 is a sensor that measures angular velocity (also called spatial angular velocity) around three axes. The angular velocity sensor 112 measures angular velocity (also called spatial angular velocity) as a physical quantity related to the movement of the foot. The angular velocity sensor 112 outputs the measured angular velocity to the feature data generation unit 12. For example, the angular velocity sensor 112 can use sensors of the vibration type, capacitive type, etc. The sensor used as the angular velocity sensor 112 is not limited to any measurement method as long as it can measure angular velocity.

[0024] Sensor 11 can be implemented, for example, by an inertial measurement device that measures acceleration and angular velocity. An example of an inertial measurement device is an IMU (Inertial Measurement Unit). An IMU includes an acceleration sensor 111 that measures acceleration in three axes and an angular velocity sensor 112 that measures angular velocity around three axes. Sensor 11 may also be implemented by an inertial measurement device such as a VG (Vertical Gyro) or AHRS (Attitude Heading). Alternatively, sensor 11 may be implemented by a GPS / INS (Global Positioning System / Inertial Navigation System). Sensor 11 may also be implemented by a device other than an inertial measurement device, as long as it can measure physical quantities related to foot movement.

[0025] Figure 3 is a conceptual diagram showing an example of how the measuring device 10 is positioned inside shoes 100 for both feet. In the example in Figure 3, the measuring device 10 is installed in a position corresponding to the underside of the arch of the foot. For example, the measuring device 10 is placed in the insole inserted into the shoe 100. For example, the measuring device 10 may be placed on the bottom surface of the shoe 100. For example, the measuring device 10 may be embedded in the body of the shoe 100. The measuring device 10 may or may not be detachable from the shoe 100. The measuring device 10 may be installed in a position other than the underside of the arch of the foot, as long as it can measure sensor data related to foot movement. The measuring device 10 may also be installed in the socks worn by the user or in anklets or other accessories worn by the user. The measuring device 10 may also be directly attached to the foot or embedded in the foot. Figure 3 shows an example in which the measuring device 10 is installed in shoes 100 for both feet. The measuring device 10 may also be installed in a single shoe 100.

[0026] In the example shown in Figure 3, a local coordinate system is set with the measuring device 10 (sensor 11) as the reference point, including the x-axis in the left-right direction, the y-axis in the front-back direction, and the z-axis in the up-down direction. The x-axis is positive to the left, the y-axis is positive backward, and the z-axis is positive upward. The orientation of the axes set on the sensor 11 may be the same for both feet or different for both feet. For example, if sensors 11 manufactured to the same specifications are placed inside both shoes 100, the up-down orientation (orientation in the Z-axis direction) of the sensors 11 placed in both shoes 100 will be the same. In that case, the three axes of the local coordinate system set for the sensor data originating from the left foot and the three axes of the local coordinate system set for the sensor data originating from the right foot will be the same for both the left and right feet.

[0027] Figure 4 is a conceptual diagram illustrating the local coordinate system (x-axis, y-axis, z-axis) set for the measuring device 10 (sensor 11) installed on the underside of the arch of the foot, and the world coordinate system (X-axis, Y-axis, Z-axis) set relative to the ground. In the world coordinate system (X-axis, Y-axis, Z-axis), with the user standing upright and facing the direction of travel, the user's lateral direction is set to the X-axis (leftward is positive), the direction of the user's back is set to the Y-axis (backward is positive), and the direction of gravity is set to the Z-axis (vertically upward is positive). Note that the example in Figure 4 conceptually shows the relationship between the local coordinate system (x-axis, y-axis, z-axis) and the world coordinate system (X-axis, Y-axis, Z-axis), and does not accurately show the relationship between the local coordinate system and the world coordinate system that changes according to the user's walking.

[0028] Figure 5 is a conceptual diagram illustrating the planes (also called body planes) set for the human body. In this embodiment, the sagittal plane divides the body into left and right halves, the coronal plane divides the body into front and back halves, and the horizontal plane divides the body horizontally. Note that, as shown in Figure 5, when the person is standing upright with the centerlines of their feet facing the direction of travel, the world coordinate system and the local coordinate system coincide. In this embodiment, rotation in the sagittal plane with the x-axis as the axis of rotation is defined as roll, rotation in the coronal plane with the y-axis as the axis of rotation is defined as pitch, and rotation in the horizontal plane with the z-axis as the axis of rotation is defined as yaw. Furthermore, the angle of rotation in the sagittal plane with the x-axis as the axis of rotation is defined as the roll angle, the angle of rotation in the coronal plane with the y-axis as the axis of rotation is defined as the pitch angle, and the angle of rotation in the horizontal plane with the z-axis as the axis of rotation is defined as the yaw angle.

[0029] As shown in Figure 2, the feature data generation unit 12 (also called a feature data generation device) includes an acquisition unit 121, a normalization unit 122, an extraction unit 123, a generation unit 125, and a transmission unit 127. For example, the feature data generation unit 12 is implemented by a microcomputer or microcontroller that performs overall control and data processing of the measurement device 10. For example, the feature data generation unit 12 includes a CPU (Central Processing Unit), RAM (Random Access Memory), ROM (Read Only Memory), flash memory, etc. The feature data generation unit 12 controls the acceleration sensor 111 and the angular velocity sensor 112 to measure angular velocity and acceleration. For example, the feature data generation unit 12 may be implemented on the side of a portable terminal (not shown) carried by the subject (user).

[0030] The acquisition unit 121 acquires acceleration in three axes from the acceleration sensor 111. The acquisition unit 121 also acquires angular velocity around the three axes from the angular velocity sensor 112. For example, the acquisition unit 121 performs analog-to-digital conversion (AD conversion) on the acquired physical quantities (analog data) such as angular velocity and acceleration. Note that the physical quantities (analog data) measured by the acceleration sensor 111 and the angular velocity sensor 112 may be converted to digital data by the acceleration sensor 111 and the angular velocity sensor 112, respectively. The acquisition unit 121 outputs the converted digital data (also called sensor data) to the normalization unit 122. The acquisition unit 121 may be configured to store the sensor data in a storage unit (not shown). The sensor data includes at least the acceleration data converted to digital data and the angular velocity data converted to digital data. The acceleration data includes acceleration vectors in three axes. The angular velocity data includes angular velocity vectors around the three axes. The acquisition time of the acceleration data and angular velocity data is associated with each data set. The acquisition unit 121 may also apply corrections to the acceleration data and angular velocity data, such as correction for mounting errors, temperature, and linearity.

[0031] The normalization unit 122 acquires sensor data from the acquisition unit 121. The normalization unit 122 extracts time-series data equivalent to one walking cycle (also called walking waveform data) from the time-series data of acceleration in the three axes and angular velocity around the three axes included in the sensor data.

[0032] Figure 6 is a conceptual diagram illustrating gait events detected in a single gait cycle based on the right foot. The horizontal axis of Figure 6 represents the gait cycle normalized with one gait cycle of the right foot set as 100 percent (%). The point when the right heel touches the ground is considered the starting point (0%), and the point when the right heel touches the ground again is considered the ending point (100%). Each of the multiple timings included in one gait cycle is a gait phase. One gait cycle of one foot is broadly divided into the stance phase and the swing phase. In the example in Figure 6, the gait cycle is normalized so that the stance phase accounts for 60% and the swing phase accounts for 40%. The stance phase is subdivided into early stance T1, mid-stance T2, late stance T3, and early swing T4. The swing phase is subdivided into early swing T5, mid-swing T6, and late swing T7. The gait waveform for one gait cycle does not necessarily have to start from the moment the heel touches the ground. For example, the starting point of the gait waveform for one gait cycle may be set to the middle of the stance phase.

[0033] Walking event E1 represents heel contact (HC), the beginning of a single step cycle. Heel contact occurs when the heel of the right foot, which was off the ground during the swing phase, lands on the ground. Walking event E2 represents opposite toe off (OTO). Opposite toe off occurs when the toes of the left foot leave the ground while the sole of the right foot remains in contact with the ground. Walking event E3 represents heel rise (HR). Heel rise occurs when the heel of the right foot lifts off the ground while the sole of the right foot remains in contact with the ground. Walking event E4 represents opposite heel contact (OHC). Opposite heel contact occurs when the heel of the left foot, which was off the ground during the swing phase of the left foot, lands on the ground. Walking event E5 represents toe-off (TO). Toe-off is the event where the toes of the right foot leave the ground while the sole of the left foot remains in contact with the ground. Walking event E6 represents foot-adjacent (FA). Foot-adjacent is the event where the left and right feet cross while the sole of the left foot remains in contact with the ground. Walking event E7 represents tibia vertical (TV). Tibia vertical is the event where the tibia of the right foot becomes nearly perpendicular to the ground while the sole of the left foot remains in contact with the ground. Walking event E8 represents heel strike (HS), the end of one walking cycle. Walking event E8 corresponds to the end of the walking cycle that began with walking event E1, and also to the beginning of the next walking cycle.

[0034] Figure 7 illustrates an example of detecting heel strike (HC) and toe-off (TO) from time-series data (solid line) of forward acceleration (Y-direction acceleration). The timing of heel strike (HC) is the timing of the minimum peak immediately following the maximum peak in the time-series data of forward acceleration (Y-direction acceleration). The maximum peak that serves as a marker for the timing of heel strike (HC) corresponds to the maximum peak of the gait waveform data for one step cycle. The interval between consecutive heel strike (HC) is one step cycle. The timing of toe-off (TO) is the timing of the rise of the maximum peak that appears after the stance phase period in the time-series data of forward acceleration (Y-direction acceleration) where no fluctuations are observed. Figure 7 also shows time-series data (dashed line) of roll angle (angular velocity around the X-axis). The midpoint between the timing of the minimum roll angle and the timing of the maximum roll angle corresponds to the mid-stance phase. For example, parameters such as walking speed, stride length, circumference, internal / external rotation, and plantarflexion / dorsiflexion (also called gait parameters) can be determined based on the mid-stance phase.

[0035] The normalization unit 122 normalizes the time of the extracted walking waveform data for one walking cycle to a walking cycle of 0-100% (also called first normalization). Timings such as 1% and 10% included in the 0-100% walking cycle are also called walking phases. Furthermore, the normalization unit 122 normalizes the walking waveform data for one walking cycle that has been first normalized so that the stance phase accounts for 60% and the swing phase accounts for 40% (also called second normalization). The stance phase is the period when at least a part of the sole of the foot is in contact with the ground. The swing phase is the period when the sole of the foot is off the ground. By second normalizing the walking waveform data, the discrepancy in the walking phase from which features are extracted can be reduced.

[0036] Figure 8 illustrates an example of gait waveform data normalized by the normalization unit 122. The normalization unit 122 detects heel strike (HC) and toe-off (TO) from time-series data of acceleration in the direction of travel (Y-direction acceleration). The normalization unit 122 extracts the interval between consecutive heel strikes (HC) as gait waveform data for one step cycle. Through first normalization, the normalization unit 122 converts the horizontal axis (time axis) of the gait waveform data for one step cycle into a gait cycle of 0-100%. In Figure 8, the gait waveform data after first normalization is shown by a dashed line. In the gait waveform data after first normalization (dashed line), the timing of toe-off (TO) is shifted from 60%.

[0037] In the example shown in Figure 8, the normalization unit 122 normalizes the section from heel strike (HC) at 0% of the walking phase to toe-off (TO) following that heel strike to 60%. The normalization unit 122 also normalizes the section from toe-off (TO) to heel strike (HC) at 100% of the walking phase following that toe-off to 60% to 100%. As a result, the walking waveform data for one walking cycle is normalized into a section where the walking cycle is 0-60% (stance phase) and a section where the walking cycle is 60-100% (swing phase). Figure 8 shows the walking waveform data after the second normalization as a solid line. In the walking waveform data after the second normalization (solid line), the timing of toe-off (TO) coincides with 60%.

[0038] Figures 7 and 8 show an example of extracting and normalizing walking waveform data for one step cycle based on the acceleration in the direction of travel (Y-direction acceleration). The normalization unit 122 extracts and normalizes walking waveform data for one step cycle in accordance with the walking cycle of the acceleration in the direction of travel (Y-direction acceleration) for accelerations and angular velocities other than the acceleration in the direction of travel (Y-direction acceleration). Alternatively, the normalization unit 122 may generate time-series data of angles around the three axes by integrating the time-series data of angular velocities around the three axes. In that case, the normalization unit 122 also extracts and normalizes walking waveform data for one step cycle in accordance with the walking cycle of the acceleration in the direction of travel (Y-direction acceleration) for angles around the three axes.

[0039] The normalization unit 122 may extract and normalize walking waveform data for one step cycle based on acceleration / angular velocity other than the acceleration in the direction of travel (Y direction acceleration) (diagram omitted). For example, the normalization unit 122 may detect heel strike (HC) and toe-off (TO) from time-series data of vertical acceleration (Z direction acceleration). The timing of heel strike (HC) is the timing of a steep minimum peak that appears in the time-series data of vertical acceleration (Z direction acceleration). At the timing of the steep minimum peak, the value of vertical acceleration (Z direction acceleration) is approximately 0. The minimum peak that serves as a marker for the timing of heel strike (HC) corresponds to the smallest peak in the walking waveform data for one step cycle. The interval between consecutive heel strike (HC) is one step cycle. The timing of toe-off (TO) is the inflection point in the time-series data of vertical acceleration (Z-direction acceleration) where it gradually increases after passing through a period of small fluctuation following a maximum peak immediately after heel strike (HC). The normalization unit 122 may also extract and normalize walking waveform data for one step cycle based on both forward acceleration (Y-direction acceleration) and vertical acceleration (Z-direction acceleration). Alternatively, the normalization unit 122 may extract and normalize walking waveform data for one step cycle based on accelerations other than forward acceleration (Y-direction acceleration) and vertical acceleration (Z-direction acceleration), such as angular velocity or angle.

[0040] The extraction unit 123 acquires walking waveform data for one walking cycle that has been normalized by the normalization unit 122. From the walking waveform data for one walking cycle, the extraction unit 123 extracts features used to estimate an index value indicating the knee condition. For example, based on pre-set conditions, the extraction unit 123 extracts features for each walking phase cluster from walking phase clusters that integrate temporally consecutive walking phases. A walking phase cluster includes at least one walking phase. A walking phase cluster may also include a single walking phase. The walking waveform data and walking phases from which features used to estimate an index value indicating the knee condition are extracted will be described later.

[0041] The generation unit 125 obtains feature quantities (first features) extracted from each walking phase that constitutes the walking phase cluster. The generation unit 125 applies a feature construct formula to the obtained first features to generate feature quantities (second features) for each walking phase cluster. The feature construct formula is a pre-set calculation formula for generating feature quantities (second features) for each walking phase cluster. For example, the feature construct formula is a calculation formula related to arithmetic operations. For example, the second features calculated using the feature construct formula are the integral mean, arithmetic mean, slope, and variability of the first features in each walking phase included in the walking phase cluster. For example, the extraction unit 123 applies a calculation formula to calculate the slope and variability of the first features extracted from each walking phase that constitutes the walking phase cluster as the feature construct formula. For example, if the walking phase cluster consists of a single walking phase, the slope and variability cannot be calculated, so a feature construct formula that calculates the integral mean or arithmetic mean should be used.

[0042] Furthermore, the generation unit 125 calculates parameters related to gait (also called gait parameters). The generation unit 125 calculates gait parameters using feature quantities derived from gait waveform data. For example, the generation unit 125 calculates stride length, maximum dorsiflexion (maximum dorsiflexion), proportion of stance phase in one step cycle, proportion of swing phase in one step cycle, maximum toe height (maximum toe height), and stride time as gait parameters. The gait parameters may also be calculated by the index value estimation device 13.

[0043] Stride length corresponds to the distance traveled in the horizontal plane during the interval from the heel strike timing, which is the starting point of a one-step cycle, to the heel strike timing, which is the ending point. For example, the generation unit 125 calculates the stride length as the distance between the starting point and the ending point of the trajectory in the horizontal plane, which is obtained by double-integrating the spatial acceleration. Maximum dorsiflexion (maximum dorsiflexion) corresponds to the maximum angle of the sole of the foot with respect to the horizontal plane. For example, the generation unit 125 calculates the maximum dorsiflexion as the spatial angle obtained by integrating the spatial angular velocity. The proportion of the stance phase is the value obtained by dividing the period from the heel strike timing, which is the starting point of a one-step cycle, to the toe-off timing by the duration of the one-step cycle. In the case of second normalization, the proportion of the stance phase is 0.6 (60%). The proportion of the swing phase is the value obtained by dividing the period from the toe-off timing to the heel strike timing, which is the ending point of a one-step cycle, by the duration of the one-step cycle. In the case of second normalization, the proportion of the swing phase is 0.4 (40%). The maximum toe height (maximum toe height) is the maximum vertical height. For example, the generation unit 125 calculates the maximum toe height as the maximum value of the vertical height obtained by double-integrating the vertical acceleration. Stride time corresponds to the time from the heel strike timing, which is the starting point of one step cycle, to the heel strike timing, which is the ending point. For example, the generation unit 125 calculates the stride time by dividing the stride length by the average value of the forward velocity in one step cycle (average walking speed), which is obtained by integrating the forward acceleration. The above calculation method is just one example and does not limit the method of calculating gait parameters.

[0044] The transmitting unit 127 outputs feature data for each walking phase cluster generated by the generating unit 125. The transmitting unit 127 transmits the generated feature data for each walking phase cluster to the index value estimation device 13 that uses that feature data. For example, the transmitting unit 127 transmits the feature data to the data relay device 15 via wireless communication. For example, the transmitting unit 127 is configured to transmit the feature data to the data relay device 15 via a wireless communication function (not shown) that conforms to standards such as Bluetooth® or WiFi®. Note that the communication function of the transmitting unit 127 may conform to standards other than Bluetooth® or WiFi®.

[0045] [Indicator Value Estimation Device] Figure 9 is a block diagram showing an example of the configuration of the index value estimation device 13. The index value estimation device 13 includes a data acquisition unit 131, a storage unit 132, an estimation unit 133, and an output unit 135.

[0046] The data acquisition unit 131 receives feature data from the measurement device 10. The data acquisition unit 131 outputs the received feature data to the estimation unit 133. The data acquisition unit 131 communicates with the transmission unit 127 of the measurement device 10 using a common communication method. The data acquisition unit 131 receives feature data from the measurement device 10 via wireless communication. For example, the data acquisition unit 131 is configured to receive feature data from the measurement device 10 via a wireless communication function (not shown) conforming to standards such as Bluetooth® or WiFi®. The communication function of the data acquisition unit 131 may conform to standards other than Bluetooth® or WiFi®. The data acquisition unit 131 may also be configured to receive feature data from the measurement device 10 via a wired connection such as a cable.

[0047] The memory unit 132 stores an estimation model that estimates index values ​​indicating the state of the knee using feature data extracted from gait waveform data. The memory unit 132 stores an estimation model that has learned the relationship between feature data related to the knee joint flexion angle of multiple subjects and index values ​​indicating the state of the knee. For example, the memory unit 132 stores an estimation model that has been learned for multiple subjects to estimate parameters related to the knee joint flexion angle. Details of the parameters related to the knee joint flexion angle will be described later.

[0048] The estimation model can be stored in the memory unit 132 at times such as when the product is shipped from the factory or when the estimation system is calibrated before a user uses it. For example, the system may be configured to use an estimation model stored in a storage device such as an external server. In that case, the system should be configured to use the estimation model via an interface (not shown) connected to that storage device.

[0049] The estimation unit 133 acquires feature data from the data acquisition unit 131. Using the acquired feature data, the estimation unit 133 estimates parameters related to the knee joint flexion angle as index values ​​indicating the state of the knee. The estimation unit 133 inputs the feature data into the estimation model stored in the memory unit 132. The estimation unit 133 outputs estimation results corresponding to the index values ​​indicating the state of the knee (parameters related to the knee joint flexion angle) output from the estimation model. When using an estimation model stored in an external storage device built on a cloud or server, the estimation unit 133 is configured to use the estimation model via an interface (not shown) connected to that storage device.

[0050] The output unit 135 outputs the estimation results of index values ​​(parameters related to knee joint flexion angle) indicating the knee condition, calculated by the estimation unit 133. For example, the output unit 135 displays the estimated results of the index values ​​indicating the knee condition on the screen of the subject's (user's) mobile device. For example, the output unit 135 outputs the estimation results to an external system that uses the estimation results. There are no particular limitations on the use of the index values ​​indicating the knee condition output from the index value estimation device 13.

[0051] For example, the index value estimation device 13 is connected to an external system built on a cloud or server via a mobile terminal (not shown) carried by the subject (user). The mobile terminal (not shown) is a portable communication device. For example, the mobile terminal is a portable communication device with communication functions such as a smartphone, smartwatch, or mobile phone. For example, the index value estimation device 13 is connected to the mobile terminal via wireless communication. For example, the index value estimation device 13 is connected to the mobile terminal via wireless communication function (not shown) conforming to standards such as Bluetooth® or WiFi®. The communication function of the index value estimation device 13 may conform to standards other than Bluetooth® or WiFi®. For example, the index value estimation device 13 may be connected to the mobile terminal via a wired connection such as a cable. The estimation result of the index value indicating the knee condition may be used by an application installed on the mobile terminal. In that case, the mobile terminal performs processing using the estimation result by application software installed on the mobile terminal.

[0052] [Parameters related to knee joint flexion angle] Next, we will explain the correlation between parameters related to knee joint flexion angle and feature data, along with the verification results. Below, we will present an example of verification conducted on 72 subjects (36 males and 36 females). In the following example, we verified the correlation between measured and estimated values ​​of parameters related to knee joint flexion angle during walking. In the following verification, subjects wearing smart apparel and shoes equipped with a measurement device 10 walked two round trips along a 5m straight path. The measured values ​​were obtained by measuring the knee flexion joint angle of subjects wearing smart apparel using a motion capture method. The predicted values ​​were estimated using sensor data measured simultaneously with the measurement of the measured values ​​by the measurement device 10 mounted on the shoes worn by the subjects. Below, the correlation between measured and estimated values ​​is verified using the value of the intraclass correlation coefficient (ICC).

[0053] Figure 10 is a graph showing an example of time-series data of knee joint flexion angle during one gait cycle. The horizontal axis of Figure 10 represents the gait cycle, starting at the timing of heel strike and ending at the timing of the next heel strike. The gait cycle is normalized to 0-100%. Furthermore, the gait cycle is normalized so that the stance phase accounts for 60% of the period and the swing phase accounts for 40%.

[0054] As shown in Figure 10, the time-series data of knee joint flexion angle during a single step cycle shows two peaks. A dip appears between the two peaks. The first peak appears during the transition period from early stance T1 to mid-stance T2. The timing of the first peak roughly coincides with the timing of opposite foot toe-off (OTO). The second peak appears during the transition period from early swing T5 to mid-swing T6. The timing of the second peak roughly coincides with the timing of foot crossing (FA).

[0055] Figure 10 shows parameters related to knee joint flexion angle as an example of indicator values ​​that show the condition of the knee. In this embodiment, an example is given in which the parameters related to knee joint flexion angle are estimated as a first angle parameter F1, a second angle parameter F2, a third angle parameter F3, a fourth angle parameter F4, a gait cycle parameter G, and a time parameter T.

[0056] Non-patent document 4 describes Stiff-knee Gait, a phenomenon in which the knee joint flexion angle decreases during walking in patients with cerebral palsy (Non-patent document 4: SR Goldberg, et al., Journal of Biomechanics, 37, (2004), pp.1189-1196). Stiff-knee Gait is defined as "limping characterized by a decrease in the knee joint flexion angle during the swing phase." In recent years, the term Stiff-knee Gait has also been used to describe abnormal gait caused by knee joint diseases. Stiff-knee Gait increases the risk of falls due to toe catching and reduces walking speed / walking energy efficiency. Osteoarthritis of the knee can also cause a decrease in the knee joint flexion angle. The presence or absence of Stiff-knee Gait may be used as a parameter to diagnose diseases such as osteoarthritis of the knee. Non-patent document 4 reports that if the knee flexion speed is insufficient when the toes leave the ground, it can lead to a stiff-knee gait, potentially reducing the knee joint flexion angle during the swing phase.

[0057] The first angle parameter F1 is the value obtained by subtracting the knee joint flexion angle at the time of the trough between the two peaks from the knee joint flexion angle at the time of the first peak. When there is a knee disorder, the trough between the two peaks tends to become unclear. Therefore, when there is a knee disorder, the first angle parameter F1 becomes smaller.

[0058] The second angle parameter F2 is the value obtained by subtracting the knee joint flexion angle at the time of the trough between the two peaks from the knee joint flexion angle at the time of the second peak. When there is a knee disorder, the trough between the two peaks tends to become unclear. Therefore, when there is a knee disorder, the second angle parameter F2 becomes smaller.

[0059] The third angle parameter F3 is the value obtained by subtracting the knee joint flexion angle at the time of toe-off from the knee joint flexion angle at the time of the second peak. When there is a knee disorder, the knee joint flexion angle tends to decrease. Therefore, when there is a knee disorder, the third angle parameter F3 will be small.

[0060] The fourth angle parameter, F4, is the knee joint flexion angle at the timing of the second peak. When there is a knee disorder, the knee joint flexion angle during the swing phase tends to decrease. Therefore, when there is a knee disorder, the fourth angle parameter F4 will be smaller.

[0061] The gait cycle parameter G is the time distance (gait cycle) from the timing of toe-off to the timing of the second peak. When there is a knee disease, the knee velocity at the timing of toe-off tends to decrease. Therefore, when there is a knee disease, the gait cycle parameter G becomes larger.

[0062] The time parameter T is the time from the moment of toe-off to the moment of the second peak. When there is a knee disorder, the knee velocity at the moment of toe-off tends to decrease. Therefore, when there is a knee disorder, the time parameter T becomes larger.

[0063] Figures 11-15 are correspondence tables summarizing the features used to estimate parameters related to knee joint flexion angle. The correspondence tables in Figures 11-15 associate the feature number, the source of feature extraction (gait waveform data and gait parameters), and the gait phase (%) included in the gait phase cluster. The features used to estimate parameters related to knee joint flexion angle were selected based on the correlation between measured and estimated values. The features used in the estimation of knee joint flexion angle parameters listed below are examples only and do not limit the features that can be used to estimate parameters related to knee joint flexion angle.

[0064] Figure 11 is a correspondence table summarizing the features used to estimate the first angular parameter F1. Features F1-1 to F11 were used to estimate the first angular parameter F1. Feature F1-1 was extracted from 94% of the walking phase of the walking waveform data Ax, which relates to the time-series data of lateral acceleration (X-direction acceleration). Feature F1-2 was extracted from the 79-81% section of the walking phase of the walking waveform data Ay, which relates to the time-series data of forward acceleration (Y-direction acceleration). Feature F1-3 was extracted from the 1%, 33%, and 43% sections of the walking phase of the walking waveform data Az, which relates to the time-series data of vertical acceleration (Z-direction acceleration). Feature F1-4 was extracted from the 39-40% section of the walking phase of the walking waveform data Gy, which relates to the time-series data of angular velocity in the coronal plane (around the Y-axis). Features F1-5 were extracted from the 62-63% walking phase of the walking waveform data Gz, which relates to time-series data of angular velocity in the horizontal plane (around the Z axis). Features F1-6 were extracted from the 68-72% and 88-93% walking phases of the walking waveform data Ex, which relates to time-series data of angle (postural angle) in the sagittal plane (around the X axis). Features F1-7 were extracted from the 6-21% and 23-28% walking phases of the walking waveform data Ey, which relates to time-series data of angle (postural angle) in the sagittal plane (around the Y axis). Feature F1-8 is the stride length included in the gait parameters. Feature F1-9 is the maximum dorsiflexion (maximum dorsiflexion) included in the gait parameters. Feature F1-10 is the proportion of the stance phase in one step cycle, included in the gait parameters. Feature F1-11 is the proportion of the swing phase in a single step cycle, included in the gait parameters. The intraclass correlation coefficient (ICC) between the measured and estimated values ​​of the first angular parameter F1 was 0.4893.

[0065] For example, an estimation model that estimates the first angle parameter F1 outputs the first angle parameter F1, which is an index value indicating the state of the knee, in response to the input of features F1-1 to F11. Such an estimation model is generated by training with training data in which the features F1-1 to F11 used to estimate the first angle parameter F1 are used as explanatory variables and the first angle parameter F1 is the target variable.

[0066] Figure 12 is a correspondence table summarizing the features used to estimate the second angular parameter F2. Features F2-1 to F2-8 were used to estimate the second angular parameter F2. Feature F2-1 was extracted from 93% of the walking phase of the walking waveform data Ax, which relates to the time-series data of lateral acceleration (X-direction acceleration). Feature F2-2 was extracted from 12% and the 78-84% section of the walking phase of the walking waveform data Ay, which relates to the time-series data of forward acceleration (Y-direction acceleration). Feature F2-3 was extracted from 25-26% of the walking phase of the walking waveform data Az, which relates to the time-series data of vertical acceleration (Z-direction acceleration). Feature F2-4 was extracted from the 70% section of the walking phase of the walking waveform data Gy, which relates to the time-series data of angular velocity in the coronal plane (around the Y-axis). Feature F2-5 was extracted from the gait waveform data Ex, specifically from the 38-44% and 63-86% intervals of the gait phase, for time-series data of angles (postural angles) in the sagittal plane (around the X-axis). Feature F2-6 was extracted from the 9-11% interval of the gait waveform data Ez, specifically from time-series data of angles (postural angles) in the horizontal plane (around the Z-axis). Feature F2-7 is the maximum toe height (maximum toe height) included in the gait parameters. Feature F2-8 is the stride time included in the gait parameters. The intraclass correlation coefficient (ICC) between the measured and estimated values ​​of the second angular parameter F2 was 0.4732.

[0067] For example, an estimation model that estimates the second angle parameter F2 outputs the second angle parameter F2, which is an index value indicating the state of the knee, in response to the input of features F2-1 to F2-8. Such an estimation model is generated by training with training data in which the features F2-1 to F2-8 used to estimate the second angle parameter F2 are used as explanatory variables and the second angle parameter F2 is the target variable.

[0068] Figure 13 is a correspondence table summarizing the features used to estimate the third angular parameter F3. Features F3-1 and F3-2 were used to estimate the third angular parameter F3. Feature F3-1 was extracted from the walking phases of 33% and 75-77% of the walking waveform data Az, which relates to the time-series data of vertical acceleration (Z-direction acceleration). Feature F3-2 was extracted from the walking phase interval of 52-82% of the walking waveform data Ex, which relates to the time-series data of angles (postural angles) in the sagittal plane (around the X-axis). The intraclass correlation coefficient (ICC) between the measured and estimated values ​​of the third angular parameter F3 was 0.5944.

[0069] For example, an estimation model that estimates the third angle parameter F3 outputs the third angle parameter F3, which is an index value indicating the state of the knee, in response to the input of features F3-1 to F3-2. Such an estimation model is generated by training with training data in which the features F3-1 to F3-2 used to estimate the third angle parameter F3 are used as explanatory variables and the third angle parameter F3 is the target variable.

[0070] Figure 14 is a correspondence table summarizing the features used to estimate the fourth angular parameter F4. Features F4-1 and F4-2 were used to estimate the fourth angular parameter F4. Feature F4-1 was extracted from the 68% walking phase of the walking waveform data Ax, which relates to the time series data of lateral acceleration (X-direction acceleration). Feature F4-2 was extracted from the 75-86% walking phase interval of the walking waveform data Ey, which relates to the time series data of angle (postural angle) in the sagittal plane (around the Y-axis). The intraclass correlation coefficient (ICC) between the measured value and the estimated value of the fourth angular parameter F4 was 0.3345.

[0071] For example, an estimation model that estimates the fourth angle parameter F4 outputs the fourth angle parameter F4, which is an index value indicating the state of the knee, in response to the input of features F4-1 to F4-2. Such an estimation model is generated by training with training data in which the features F4-1 to F4-2 used to estimate the fourth angle parameter F4 are used as explanatory variables and the fourth angle parameter F4 is the target variable.

[0072] Figure 15 is a correspondence table summarizing the features used to estimate the gait cycle parameter G and the time parameter T. Features G-1 to G-3 were used to estimate the gait cycle parameter G. Features T-1 to G-3 were used to estimate the time parameter T. Features G-1 and T-1 were extracted from 87% of the gait phase of the gait waveform data Ax, which relates to time-series data of lateral acceleration (X-direction acceleration). Features G-2 and T-2 were extracted from the 76-78% section of the gait phase of the gait waveform data Ez, which relates to time-series data of angles (postural angles) in the horizontal plane (around the Z-axis). Features G-3 and T-3 were extracted from the 1-3% and 67-83% sections of the gait phase of the gait waveform data Ex, which relates to time-series data of angles (postural angles) in the sagittal plane (around the X-axis). The intraclass correlation coefficient (ICC) between the measured and estimated values ​​of the gait cycle parameter G was 0.4818. The intraclass correlation coefficient (ICC) between the measured and estimated values ​​of the time parameter T was 0.7122.

[0073] For example, an estimation model that estimates the gait cycle parameter G outputs the gait cycle parameter G, which is an index value indicating the state of the knee, in response to the input of features G-1 to G-3. Such an estimation model is generated by training using training data in which the features G-1 to G-3 used to estimate the gait cycle parameter G are used as explanatory variables and the gait cycle parameter G is the target variable.

[0074] For example, an estimation model that estimates the time parameter T outputs the time parameter T, which is an index value indicating the state of the knee, in response to the input of features T-1 to T-3. Such an estimation model is generated by training with training data in which the features T-1 to T-3 used to estimate the time parameter T are used as explanatory variables and the time parameter T is the target variable.

[0075] If estimation results regarding parameters related to knee joint flexion angle are output in response to the input of feature data, the estimation results of the estimation model are not limited. For example, the memory unit 132 stores an estimation model that estimates parameters related to knee joint flexion angle using a multiple regression prediction method. For example, the memory unit 132 stores coefficients (weights) that are accumulated on individual feature data. The coefficients (weights) stored in the memory unit 132 are accumulated on the corresponding feature data. The sum of the feature data on which the coefficients (weights) are accumulated corresponds to the parameters related to knee joint flexion angle.

[0076] (operation) Next, an example of the operation of the estimation system 1 will be explained with reference to the diagram. Here, the measurement device 10 and the index value estimation device 13 included in the estimation system 1 will be explained individually. Regarding the measurement device 10, the operation of the feature data generation unit 12 included in the measurement device 10 will be explained.

[0077] [Measuring device] Figure 16 is a flowchart illustrating the operation of the feature data generation unit 12 included in the measurement device 10. In the explanation following the flowchart in Figure 16, the feature data generation unit 12 will be described as the main operator.

[0078] In Figure 16, first, the feature data generation unit 12 acquires time-series data of sensor data related to foot movement (step S101).

[0079] Next, the feature data generation unit 12 extracts walking waveform data for one step cycle from the time-series data of the sensor data (step S102). The feature data generation unit 12 detects heel strike and toe-off from the time-series data of the sensor data. The feature data generation unit 12 extracts the time-series data of the interval between consecutive heel strikes as walking waveform data for one step cycle.

[0080] Next, the feature data generation unit 12 normalizes the extracted walking waveform data for one walking cycle (step S103). The feature data generation unit 12 normalizes the walking waveform data for one walking cycle to a walking cycle of 0 to 100% (first normalization). Furthermore, the feature data generation unit 12 normalizes the ratio of the stance phase to the swing phase of the walking waveform data for one walking cycle that has been first normalized to 60:40 (second normalization).

[0081] Next, the feature data generation unit 12 extracts features from the normalized gait waveform, specifically from the gait phase used to estimate parameters related to the knee joint flexion angle (step S104). The feature data generation unit 12 then extracts features to be input into a pre-constructed estimation model.

[0082] Next, the feature data generation unit 12 generates feature data for each walking phase cluster using the extracted features (step S105).

[0083] Next, the feature data generation unit 12 integrates the features for each walking phase cluster to generate feature data for one step cycle (step S106).

[0084] Next, the feature data generation unit 12 outputs the generated feature data to the index value estimation device 13 (step S107).

[0085] [Indicator Value Estimation Device] Figure 17 is a flowchart illustrating the operation of the index value estimation device 13. In the explanation following the flowchart in Figure 17, the index value estimation device 13 will be described as the main operating component.

[0086] In Figure 17, first, the index value estimation device 13 acquires feature data used to estimate parameters related to the knee joint flexion angle (step S131).

[0087] Next, the index value estimation device 13 inputs the acquired feature data into an estimation model that estimates parameters related to the knee joint flexion angle (step S132).

[0088] Next, the index value estimation device 13 estimates parameters related to the user's knee joint flexion angle according to the output (estimated value) from the estimation model (step S133).

[0089] Next, the index value estimation device 13 outputs information about the estimated parameters (step S134). For example, the parameter related to the knee joint flexion angle is output to a terminal device (not shown) carried by the user. For example, the parameter related to the knee joint flexion angle is output to a system that performs processing using the parameter.

[0090] (Examples of application) Next, an example of application according to this embodiment will be described with reference to the drawings. In the following example, an example is shown in which the function of an index value estimation device 13 installed on a mobile terminal carried by the user estimates information regarding an index value indicating the condition of the knee, using feature data measured by a measuring device 10 placed on the shoe 100.

[0091] Figure 18 is a conceptual diagram showing an example of displaying the estimation results by the index value estimation device 13 on the screen of a mobile terminal 160 carried by a user walking while wearing shoes 100 equipped with a measuring device 10. Figure 18 is an example of displaying information on the screen of the mobile terminal 160 that corresponds to the estimation results of parameters related to the knee joint flexion angle, which are estimated using feature data corresponding to sensor data measured during the user's walking.

[0092] Figure 18 shows an example of information displayed on the screen of the mobile device 160, based on the estimated results of a parameter related to knee joint flexion angle, which is an indicator value of the knee condition. In the example in Figure 18, based on the estimated results, the information "There is a tendency for the knee joint flexion angle during the swing phase to be small" is displayed on the display of the mobile device 160. Also in the example in Figure 18, based on the estimated value of the parameter related to knee joint flexion angle, which is an indicator value of the knee condition, recommendation information "We recommend that you see a doctor at a hospital" is displayed on the display of the mobile device 160. For example, a link to the website or phone number of a hospital where a consultation is possible may be displayed on the screen of the mobile device 160. After checking the information displayed on the display of the mobile device 160, the user can go to a hospital as recommended and receive a consultation for a knee-related illness.

[0093] As described above, the estimation system of this embodiment comprises a measurement device and an index value estimation device. The measurement device comprises a sensor and a feature data generation unit. The sensor has an acceleration sensor and an angular velocity sensor. The sensor measures spatial acceleration using the acceleration sensor. The sensor measures spatial angular velocity using the angular velocity sensor. The sensor generates sensor data related to foot movement using the measured spatial acceleration and spatial angular velocity. The sensor outputs the generated sensor data to the feature data generation unit. The feature data generation unit acquires time-series data of the sensor data related to foot movement. The feature data generation unit extracts walking waveform data for one step cycle from the time-series data of the sensor data. The feature data generation unit normalizes the extracted walking waveform data. From the normalized walking waveform data, the feature data generation unit extracts features used to estimate index values ​​indicating knee condition from a walking phase cluster consisting of at least one temporally consecutive walking phase. The feature data generation unit generates feature data including the extracted features. The feature data generation unit outputs the generated feature data.

[0094] The index value estimation device comprises a data acquisition unit, a storage unit, an estimation unit, and an output unit. The data acquisition unit acquires feature data, including features used to estimate an index value indicating the user's knee condition, extracted from sensor data related to the user's foot movements. The storage unit stores an estimation model that outputs an index value corresponding to the input of feature data. The storage unit stores an estimation model that estimates parameters related to the knee joint flexion angle as an index value indicating the knee condition. The estimation unit estimates the output obtained by inputting the acquired feature data into the estimation model as an index value indicating the user's knee condition. The output unit outputs information regarding the estimated index value indicating the user's knee condition. The estimation unit estimates the parameters related to the knee joint flexion angle obtained by inputting the acquired feature data about the user into the estimation model as an index value indicating the user's knee condition.

[0095] The estimation system of this embodiment estimates an index value indicating the user's knee condition using feature quantities extracted from sensor data related to the user's foot movements. Therefore, according to this embodiment, parameters related to the knee joint flexion angle can be appropriately estimated as an index value indicating the knee condition in daily life without using a dedicated device for measuring the index value indicating the knee condition.

[0096] The knee plays a crucial role in everyday walking. Knee-related diseases such as osteoarthritis can cause pain due to arthritis and other factors that negatively impact the quality of life (QoL) in daily life. Early detection and prevention are important for these diseases. However, diagnosing these diseases requires measurements using specialized equipment and diagnosis by specialists. Therefore, it is difficult to detect and prevent these diseases early in daily life. According to the method of this embodiment, parameters related to knee joint flexion angle can be appropriately estimated using sensor data measured in daily life. If the estimated parameter values ​​deviate significantly from the healthy range, there is a possibility of a knee disease. The parameters obtained by the method of this embodiment can be used as auxiliary information for diagnosing knee diseases. In other words, according to the method of this embodiment, indicator values ​​for the early detection and prevention of knee-related diseases can be estimated in daily life.

[0097] In one embodiment of this system, the memory unit stores an estimation model generated by training using training data in which feature quantities obtained from verification of walking patterns of multiple subjects are used as explanatory variables and measured index values ​​measured from verification of walking patterns of multiple subjects are used as the objective variable. The estimation unit inputs feature data obtained about the user into the estimation model and estimates the output as an index value indicating the condition of the user's knees. According to this embodiment, by using an estimation model trained with feature quantities obtained from verification of multiple subjects and measured index values, an index value indicating the condition of the knees can be estimated from a statistical standpoint.

[0098] In one embodiment of this system, the storage unit stores an estimation model that estimates parameters related to knee joint flexion angle associated with two peaks appearing in time-series data of knee joint flexion angle for one walking cycle. The estimation unit inputs feature data acquired in accordance with the user's walking into the estimation model and estimates an index value indicating the user's knee condition according to the user's index value output from the estimation model. In this embodiment, parameters related to knee joint flexion angle are estimated in association with the peaks appearing in the time-series data of knee joint flexion angle. Therefore, according to this embodiment, it is possible to estimate an index value that better reflects the movement of the knee during walking.

[0099] In one embodiment of this system, the memory unit stores an estimation model that estimates parameters related to the knee joint flexion angle, including the temporal relationship between the timing of the peak appearing in the swing phase of the two peaks in the time-series data of the knee joint flexion angle for one walking cycle and the timing of toe-off. The estimation unit inputs feature data acquired according to the user's walking into the estimation model and estimates an index value indicating the user's knee condition according to the user's index value output from the estimation model. In this embodiment, the temporal relationship between the timing of the peak appearing in the swing phase of the time-series data of the knee joint flexion angle and the timing of toe-off is estimated as a parameter related to the knee joint flexion angle. Therefore, according to this embodiment, an index value that better reflects the movement of the knee in the swing phase can be estimated.

[0100] (Second embodiment) Next, the estimation system according to the second embodiment will be described with reference to the drawings. In this embodiment, an example is given in which the cost indicating the smoothness of knee movement is estimated as an index value indicating the state of the knee. In this embodiment, an example is given in which AJC (Angular Jerk Cost), which corresponds to the value obtained by accumulating the squared value of the angular jerk, which is the third derivative of the knee flexion angle, over a specific period included in the gait cycle, is estimated as the cost indicating the smoothness of knee movement.

[0101] (composition) Figure 19 is a block diagram showing an example of the configuration of the estimation system 2 according to this embodiment. The estimation system 2 comprises a measuring device 20 and an index value estimation device 23. The measuring device 20 has the same configuration as the measuring device 10 of the first embodiment. The index value estimation device 23 has the same configuration as the index value estimation device 13 of the first embodiment. The measuring device 20 and the index value estimation device 23 differ from the first embodiment in that they extract features used to estimate the index value indicating the state of the knee. In the following, the detailed configurations of the measuring device 20 and the index value estimation device 23 will be omitted from the explanation.

[0102] There are two main interpretations of the evaluation of angular jerk. One interpretation is that the value of angular jerk increases when greater muscle force is exerted. The other interpretation is that the value of angular jerk increases when the smoothness of movement decreases. Subjects with osteoarthritis of the knee have difficulty making appropriate kinematic responses in the initial stance phase due to factors such as knee pain and limited range of motion, which affect knee joint function and gait. It is presumed that such subjects take measures to reduce the change in angular acceleration of the knee joint by reducing the ground reaction force, thereby ensuring smooth movement in order to avoid knee pain. In this embodiment, it is assumed that the smoothness of movement increases and the angular jerk decreases in response to compensatory movements to avoid knee pain. Normally, the movement of the knee angle is not uniformly accelerated motion. However, when compensatory movements are taken to alleviate knee pain, the movement of the knee angle tends to become closer to uniformly accelerated motion.

[0103] Figure 20 is a graph showing an example of time-series data of angular jerk. In this embodiment, AJC is estimated for each of several target segments included in the 0-60% interval (stance phase) of the gait cycle. In this embodiment, AJC is estimated for each of the first segment P1, second segment P2, third segment P3, and fourth segment P4 included in the stance phase. The first segment P1 is the interval from initial contact (IC) to the load reaction period (LR). Initial contact (IC) is the timing immediately after heel strike (HC). The load reaction period (LR) is the timing when the gait cycle is approximately 15%. The second segment P2 is the interval from the load reaction period (LR) to the mid-stance MS. Mid-stance MS is the timing of the transition from mid-stance T2 to terminal stance T3. Mid-stance MS is the timing in the middle of the stance phase. The third segment P3 is the interval from mid-stance MS to terminal stance (TS). Terminal stance (TS) is the transition from the end of the stance phase (T3) to the early swing phase (T4).

[0104] Similar to the first embodiment, the measurement device 20 is mounted on the subject's shoe. The measurement device 20 measures sensor data including acceleration in three axes (spatial acceleration) and angular velocity around three axes (spatial angular velocity). The measurement device 20 normalizes the measured sensor data and extracts time-series data (also called gait waveform data) for one step cycle. The measurement device 20 extracts features used for AJC estimation from the gait waveform data. The measurement device 20 extracts features for each of the first, second, third, and fourth intervals P1, P2, P2, P3, and P4. For example, the measurement device 20 extracts features from each of the first, second, third, and fourth intervals P1, P2, P2, P3, and P4. The measurement device 20 uses the extracted features to generate feature data for each gait phase cluster. The measurement device 20 transmits the generated gait phase cluster feature data to the index value estimation device 23, which uses that feature data.

[0105] The index value estimation device 23 receives feature data from the measurement device 20. The index value estimation device 23 communicates with the measurement device 20 using a common communication method. The index value estimation device 23 stores an estimation model for estimating AJC using feature data extracted from gait waveform data. The index value estimation device 23 stores feature data related to AJC for multiple subjects and an estimation model that has learned the relationship with AJC. For example, the index value estimation device 23 stores an estimation model for estimating AJC that has been learned for multiple subjects.

[0106] The index value estimation device 23 uses the acquired feature data to estimate AJC as an index value indicating the knee condition. The index value estimation device 23 inputs the feature data into the stored estimation model. The index value estimation device 23 outputs an estimation result corresponding to the index value (AJC) indicating the knee condition output from the estimation model. When using an estimation model stored in an external storage device built on a cloud or server, the index value estimation device 23 is configured to use the estimation model via an interface (not shown) connected to that storage device.

[0107] The index value estimation device 23 outputs the estimated result of an index value (AJC) indicating the condition of the knee. For example, the index value estimation device 23 displays the estimated result of the index value indicating the condition of the knee on the screen of the subject's (user's) mobile device. For example, the index value estimation device 23 outputs the estimated result to an external system that uses the estimation result. There are no particular limitations on the use of the index value indicating the condition of the knee output from the index value estimation device 23.

[0108] Next, we will explain the correlation between the estimated AJC and feature data in each interval, along with the verification results. Below, we will present a verification example conducted on 72 subjects (36 males and 36 females), similar to the first embodiment. The features used to estimate AJC were selected based on the correlation between the measured and estimated values. The intraclass correlation coefficient (ICC) between the measured and estimated values ​​in the first interval P1 was 0.2453. In the second interval P2, the intraclass correlation coefficient (ICC) between the measured and estimated values ​​was 0.4418. In the third interval P3, the intraclass correlation coefficient (ICC) between the measured and estimated values ​​was 0.6114. In the fourth interval P4, the intraclass correlation coefficient (ICC) between the measured and estimated values ​​was 0.6185. The intraclass correlation coefficient (ICC) between the measured and estimated values ​​differed depending on the interval. In the first interval P1, the movement of the measuring device 20 was complex, and the sensor data was prone to noise. As a result, it is presumed that the measured values, estimated values, and intraclass correlation coefficient (ICC) decreased. On the other hand, in the third interval P3 and the fourth interval P4, it is presumed that the movement of the measuring device 20 was stable, and the measured values, estimated values, and intraclass correlation coefficient (ICC) were relatively good.

[0109] (operation) Next, an example of the operation of the estimation system 2 will be explained with reference to the diagram. Here, the measuring device 20 and the index value estimation device 23 included in the estimation system 2 will be explained individually.

[0110] [Measuring device] Figure 21 is a flowchart illustrating the operation of the measuring device 20. In the explanation following the flowchart in Figure 21, the measuring device 20 will be described as the main operating component.

[0111] In Figure 21, first, the measuring device 20 acquires time-series data of sensor data related to foot movement (step S201).

[0112] Next, the measuring device 20 extracts gait waveform data for one step cycle from the time-series data of the sensor data (step S202). The measuring device 20 detects heel strike and toe-off from the time-series data of the sensor data. The measuring device 20 extracts the time-series data of the interval between consecutive heel strikes as gait waveform data for one step cycle.

[0113] Next, the measuring device 20 normalizes the extracted gait waveform data for one step (step S203). The measuring device 20 normalizes the gait waveform data for one step to a gait cycle of 0 to 100% (first normalization). Furthermore, the measuring device 20 normalizes the ratio of the stance phase to the swing phase of the gait waveform data for the first normalized one step to 60:40 (second normalization).

[0114] Next, the measurement device 20 extracts features from the gait phase used for AJC estimation from the normalized gait waveform (step S204). The measurement device 20 extracts features to be input into the pre-constructed estimation model.

[0115] Next, the measurement device 20 generates feature quantities for each walking phase cluster using the extracted feature quantities (step S205).

[0116] Next, the measurement device 20 integrates the feature quantities for each walking phase cluster to generate feature quantity data for one walking cycle (step S206).

[0117] Next, the measurement device 20 outputs the generated feature data to the index value estimation device 23 (step S207).

[0118] [Indicator Value Estimation Device] Figure 22 is a flowchart illustrating the operation of the index value estimation device 23. In the explanation following the flowchart in Figure 22, the index value estimation device 23 will be described as the main operating component.

[0119] In Figure 22, first, the index value estimation device 23 acquires feature data used for estimating AJC (step S231).

[0120] Next, the index value estimation device 23 inputs the acquired feature data into an estimation model that estimates AJC (step S232).

[0121] Next, the index value estimation device 23 estimates AJC according to the output (estimated value) from the estimation model (step S233).

[0122] Next, the index value estimation device 23 outputs information about the estimated AJC (step S234). For example, the AJC is output to a terminal device (not shown) carried by the user. For example, the AJC is output to a system that performs processing using parameters.

[0123] (Examples of application) Next, an example of application according to this embodiment will be described with reference to the drawings. In the following example, an example is shown in which the function of an index value estimation device 23 installed on a mobile terminal carried by the user estimates information regarding an index value indicating the condition of the knee, using feature data measured by a measuring device 20 placed on the shoe 200.

[0124] Figure 23 is a conceptual diagram showing an example of displaying the estimation results by the index value estimation device 23 on the screen of a mobile terminal 260 carried by a user walking while wearing shoes 200 equipped with a measuring device 20. Figure 23 is an example of displaying information corresponding to the AJC estimation results, which are estimated using feature data corresponding to sensor data measured during the user's walking, on the screen of the mobile terminal 260.

[0125] Figure 23 shows an example of information displayed on the screen of the mobile device 260, based on the estimated AJC (Arm Joint Cuff), an indicator of knee condition. In the example in Figure 23, the information "Your AJC tends to be small" is displayed on the display of the mobile device 260, based on the estimated AJC. In addition, in the example in Figure 23, recommendation information based on the estimated AJC, such as "We recommend exercises to strengthen your knees. Training A is optimal. Please watch the video below," is displayed on the display of the mobile device 260. After viewing the information displayed on the display of the mobile device 260, the user can refer to the video for Training A and perform the exercises, thereby improving their knee condition.

[0126] As described above, the estimation system of this embodiment comprises a measurement device and an index value estimation device. The measurement device comprises a sensor and a feature data generation unit. The sensor has an acceleration sensor and an angular velocity sensor. The sensor measures spatial acceleration using the acceleration sensor. The sensor measures spatial angular velocity using the angular velocity sensor. The sensor generates sensor data related to foot movement using the measured spatial acceleration and spatial angular velocity. The sensor outputs the generated sensor data to the feature data generation unit. The feature data generation unit acquires time-series data of the sensor data related to foot movement. The feature data generation unit extracts walking waveform data for one step cycle from the time-series data of the sensor data. The feature data generation unit normalizes the extracted walking waveform data. From the normalized walking waveform data, the feature data generation unit extracts features used to estimate index values ​​indicating knee condition from a walking phase cluster consisting of at least one temporally consecutive walking phase. The feature data generation unit generates feature data including the extracted features. The feature data generation unit outputs the generated feature data.

[0127] The index value estimation device comprises a data acquisition unit, a storage unit, an estimation unit, and an output unit. The data acquisition unit acquires feature data, including features used to estimate an index value indicating the user's knee condition, extracted from sensor data related to the user's foot movements. The storage unit stores an estimation model that outputs an index value corresponding to the input of feature data. The storage unit stores an estimation model that estimates the cost indicating the smoothness of knee movement as an index value indicating the knee condition. The estimation unit estimates the output obtained by inputting the acquired feature data into the estimation model as an index value indicating the user's knee condition. The output unit outputs information regarding the estimated index value indicating the user's knee condition. The estimation unit estimates the cost indicating the smoothness of knee movement, obtained by inputting the acquired feature data about the user into the estimation model, as an index value indicating the user's knee condition.

[0128] The estimation system of this embodiment estimates an index value indicating the user's knee condition using feature quantities extracted from sensor data related to the user's foot movements. Therefore, according to this embodiment, without using a dedicated device to measure the index value indicating the knee condition, it is possible to appropriately estimate the cost indicating the smoothness of knee movement as an index value indicating the knee condition in daily life.

[0129] According to the method of this embodiment, the cost indicating the smoothness of knee movement can be appropriately estimated using sensor data measured in daily life. If the estimated cost value deviates significantly from the healthy range, there is a possibility of a knee disease. The parameters obtained by the method of this embodiment can be used as auxiliary information for diagnosing knee diseases. In other words, according to the method of this embodiment, indicator values ​​for early detection / prevention of knee-related diseases can be estimated in daily life.

[0130] In one embodiment of this system, the memory unit stores an estimation model that estimates the cost of smooth knee movement as an index value indicating the state of the knee for each of a plurality of intervals included in the stance phase. The estimation unit estimates the cost of smooth knee movement, obtained by inputting feature data acquired about the user for at least one of the plurality of intervals into the estimation model, as an index value indicating the state of the user's knee. In this embodiment, the cost of smooth knee movement in the stance phase is estimated as an index value indicating the state of the knee. People who walk while enduring knee pain tend to have smoother knee movement in the stance phase. Therefore, according to this embodiment, an index value that can detect users with knee abnormalities can be estimated.

[0131] (Third embodiment) Next, the index value estimation device according to the third embodiment will be described with reference to the drawings. The index value estimation device of this embodiment has a simplified configuration compared to the index value estimation device included in the estimation system of the first and second embodiments.

[0132] Figure 24 is a block diagram showing an example of the configuration of the index value estimation device 33 according to this embodiment. The index value estimation device 33 comprises a data acquisition unit 331, a storage unit 332, an estimation unit 333, and an output unit 335.

[0133] The data acquisition unit 331 acquires feature data, which includes features extracted from sensor data related to the user's foot movements and used to estimate an index value indicating the user's knee condition. The storage unit 332 stores an estimation model that outputs an index value corresponding to the input of feature data. The estimation unit 333 estimates the output obtained by inputting the acquired feature data into the estimation model as an index value indicating the user's knee condition. The output unit 335 outputs information regarding the estimated index value indicating the user's knee condition.

[0134] As described above, in this embodiment, an index value indicating the user's knee condition is estimated using feature quantities extracted from sensor data related to the user's foot movements. Therefore, according to this embodiment, an index value indicating the knee condition can be appropriately estimated in daily life without using a dedicated device for measuring the index value indicating the knee condition.

[0135] (Hardware) Here, the hardware configuration for executing the control and processing according to each embodiment of this disclosure will be explained using the information processing device 90 in Figure 25 as an example. Note that the information processing device 90 in Figure 25 is an example configuration for executing the control and processing of each embodiment and does not limit the scope of this disclosure.

[0136] As shown in Figure 25, the information processing device 90 comprises a processor 91, main memory 92, auxiliary storage 93, input / output interface 95, and communication interface 96. In Figure 25, interface is abbreviated as I / F (Interface). The processor 91, main memory 92, auxiliary storage 93, input / output interface 95, and communication interface 96 are connected to each other via a bus 98, enabling data communication. Furthermore, the processor 91, main memory 92, auxiliary storage 93, and input / output interface 95 are connected to a network such as the Internet or an intranet via the communication interface 96.

[0137] The processor 91 loads the program stored in the auxiliary storage device 93, etc., into the main memory 92. The processor 91 executes the program loaded into the main memory 92. In this embodiment, a configuration using software programs installed in the information processing device 90 is sufficient. The processor 91 executes the control and processing according to each embodiment.

[0138] The main memory 92 has an area where the program is loaded. The processor 91 loads the program stored in the auxiliary memory 93, etc., into the main memory 92. The main memory 92 is implemented by volatile memory such as DRAM (Dynamic Random Access Memory). Alternatively, non-volatile memory such as MRAM (Magnetoresistive Random Access Memory) may be configured / added as the main memory 92.

[0139] The auxiliary storage device 93 stores various data, such as programs. The auxiliary storage device 93 is implemented by a local disk such as a hard disk or flash memory. It is also possible to omit the auxiliary storage device 93 by configuring the system to store various data in the main memory 92.

[0140] The input / output interface 95 is an interface for connecting the information processing device 90 to peripheral devices, based on standards and specifications. The communication interface 96 is an interface for connecting to external systems and devices via a network such as the Internet or an intranet, based on standards and specifications. The input / output interface 95 and the communication interface 96 may be common as interfaces for connecting to external devices.

[0141] The information processing device 90 may be connected to input devices such as a keyboard, mouse, or touch panel, as needed. These input devices are used to input information and settings. When a touch panel is used as an input device, the display screen of the display device may also serve as the interface for the input device. Data communication between the processor 91 and the input device can be mediated by the input / output interface 95.

[0142] Furthermore, the information processing device 90 may be equipped with a display device for displaying information. If a display device is provided, it is preferable that the information processing device 90 is equipped with a display control device (not shown) for controlling the display of the display device. The display device can be connected to the information processing device 90 via an input / output interface 95.

[0143] Furthermore, the information processing device 90 may be equipped with a drive device. The drive device mediates between the processor 91 and the recording medium (program recording medium), such as reading data and programs from the recording medium and writing the processing results of the information processing device 90 to the recording medium. The drive device can be connected to the information processing device 90 via an input / output interface 95.

[0144] The above is an example of a hardware configuration for enabling the control and processing according to each embodiment of the present invention. Note that the hardware configuration in Figure 25 is an example of a hardware configuration for executing the control and processing according to each embodiment, and does not limit the scope of the present invention. Furthermore, a program that causes a computer to execute the control and processing according to each embodiment is also included in the scope of the present invention. Moreover, a program recording medium that records the program according to each embodiment is also included in the scope of the present invention. The recording medium can be, for example, an optical recording medium such as a CD (Compact Disc) or DVD (Digital Versatile Disc). The recording medium may also be a semiconductor recording medium such as a USB (Universal Serial Bus) memory or an SD (Secure Digital) card. Furthermore, the recording medium may also be a magnetic recording medium such as a flexible disk, or other recording media. When a program executed by a processor is recorded on a recording medium, that recording medium corresponds to a program recording medium.

[0145] The components of each embodiment may be combined in any way. Furthermore, the components of each embodiment may be implemented by software or by circuitry.

[0146] Although the present invention has been described above with reference to embodiments, the present invention is not limited to the above embodiments. Various modifications to the configuration and details of the present invention can be made that will be understood by those skilled in the art within the scope of the present invention. [Explanation of Symbols]

[0147] 1, 2 Estimation System 10, 20 Measuring devices 11 sensors 12 Feature Data Generation Unit 13, 23, 33 Indicator Value Estimator 111 Accelerometer 112 Angular velocity sensor 121 Acquisition Department 122 Normalization section 123 Extraction part 125 Generation part 127 Transmitter 131, 331 Data acquisition unit 132, 332 storage section 133, 333 Estimation part 135, 335 Output section

Claims

1. A data acquisition means for acquiring feature data, which includes feature quantities extracted from sensor data related to the user's foot movements and used to estimate an index value indicating the user's knee condition; A storage means for storing an estimation model that outputs an index value corresponding to the input of the aforementioned feature data, An estimation means that estimates the output obtained by inputting the acquired feature data into the estimation model as the index value indicating the user's knee condition, The system includes an output means that outputs information regarding the index value indicating the estimated knee condition of the user, The feature data includes features extracted from walking phase clusters, which consist of at least one temporally consecutive walking phase, in walking waveform data extracted from the sensor data for one walking cycle, normalized so that the stance phase accounts for 60 percent, the swing phase accounts for 40 percent, and the timing of toe-off coincides with 60 percent. The estimation model is an index value estimation device that is a machine learning model trained to output parameters related to the knee joint flexion angle, including the timing of the peak appearing in the swing phase of the two peaks that appear in the time series data of the knee joint flexion angle for one step period, and the timing of toe-off, in response to the input of the feature data.

2. The aforementioned storage means is The estimation model generated by learning using training data, in which the feature quantities used to estimate the index value indicating the knee condition extracted from the sensor data obtained from the walking verification of multiple subjects are used as explanatory variables, and the measured values ​​of the index value indicating the knee condition measured in the walking verification of multiple subjects are used as the dependent variable, is stored. The estimation means is, The index value estimation device according to claim 1, wherein the output obtained by inputting the feature data acquired with respect to the user into the estimation model is estimated as the index value indicating the knee condition of the user.

3. The aforementioned storage means is The estimation model, which estimates parameters related to the knee joint flexion angle as the index value indicating the condition of the knee, is stored. The estimation means is, The index value estimation device according to claim 2, wherein parameters related to the knee joint flexion angle obtained by inputting the feature data acquired with respect to the user into the estimation model are estimated as index values ​​indicating the knee condition of the user.

4. The aforementioned storage means is The estimation model is stored, which estimates parameters related to the knee joint flexion angle that appear in the time-series data of the knee joint flexion angle for one gait cycle, The estimation means is, An index value estimation device according to claim 3, which inputs the feature data acquired in accordance with the user's walking into the estimation model, and estimates the index value indicating the user's knee condition according to the index value of the user output from the estimation model.

5. The estimation means is, An index value estimation device according to claim 3, which inputs the feature data acquired in accordance with the user's walking into the estimation model, and estimates the index value indicating the user's knee condition according to the index value of the user output from the estimation model.

6. The aforementioned storage means is A second estimation model is stored to estimate the cost indicating the smoothness of knee movement as the aforementioned index value indicating the condition of the knee. The estimation means is, The index value estimation device according to claim 2, wherein the cost indicating the smoothness of knee movement, obtained by inputting the feature data acquired with respect to the user into the second estimation model, is estimated as the index value indicating the knee condition of the user.

7. The aforementioned storage means is For each of the multiple segments included in the stance phase, the second estimation model is stored, which estimates the cost indicating the smoothness of knee movement as an index value indicating the state of the knee. The estimation means is, The index value estimation device according to claim 6, wherein the cost indicating the smoothness of knee movement, obtained by inputting the feature data acquired for the user into the second estimation model for at least one of the plurality of intervals, is estimated as the index value indicating the knee condition of the user.

8. An index value estimation device according to any one of claims 1 to 7, It comprises a measuring device installed on the user's footwear, which is the target of estimating an index value indicating the condition of the knee, The aforementioned measuring device is A sensor that measures spatial acceleration and spatial angular velocity, generates sensor data related to foot movement using the measured spatial acceleration and spatial angular velocity, and outputs the generated sensor data, An estimation system comprising: a feature data generation means that acquires time-series data of sensor data including gait characteristics; extracts walking waveform data for one walking cycle from the time-series data of sensor data; normalizes the extracted walking waveform data; extracts features used for estimating the index value indicating the knee state from the normalized walking waveform data from a walking phase cluster consisting of at least one temporally consecutive walking phase; generates feature data including the extracted features; and outputs the generated feature data to the index value estimation device.

9. Computers Feature data is obtained, which includes features extracted from sensor data related to the user's foot movements and used to estimate an index value indicating the user's knee condition. The acquired feature data is input to an estimation model that outputs an index value corresponding to the input of the feature data. The output obtained by inputting it into the estimation model is estimated as the index value indicating the user's knee condition. Output information regarding the index value indicating the estimated knee condition of the user, The feature data includes features extracted from walking phase clusters, which consist of at least one temporally consecutive walking phase, in walking waveform data extracted from the sensor data for one walking cycle, normalized so that the stance phase accounts for 60 percent, the swing phase accounts for 40 percent, and the timing of toe-off coincides with 60 percent. The estimation model is a machine learning model that, in response to the input of the feature data, outputs parameters related to the knee joint flexion angle, including the timing of the peak appearing in the swing phase of the two peaks that appear in the time series data of the knee joint flexion angle for one step period, and the timing of toe-off.

10. A process for obtaining feature data, which includes features extracted from sensor data related to the user's foot movements and used to estimate an index value indicating the user's knee condition, The estimation model outputs an index value corresponding to the input of the feature data, and the process involves inputting the acquired feature data into the estimation model. The process involves estimating the output obtained by inputting it into the estimation model as the index value indicating the user's knee condition, The computer is made to perform a process that outputs information regarding the index value indicating the estimated knee condition of the user, The feature data includes features extracted from walking phase clusters, which consist of at least one temporally consecutive walking phase, in walking waveform data extracted from the sensor data for one walking cycle, normalized so that the stance phase accounts for 60 percent, the swing phase accounts for 40 percent, and the timing of toe-off coincides with 60 percent. The estimation model is a machine learning model program that, in response to the input of the feature data, outputs parameters related to the knee joint flexion angle, including the temporal relationship between the timing of the peak appearing in the swing phase of the two peaks that appear in the time-series data of the knee joint flexion angle for one step period and the timing of toe-off.

Images