Gait analysis method and apparatus

Abstract

By using the helical axis in the knee joint movement of subjects during walking, we perform gait analysis that can be used to detect early-stage osteoarthritis of the knee. [Solution] A gait analysis method comprising the steps of: calculating the helical axis of the knee joint based on motion data of the subject's lower limbs during walking; extracting the flexion phase and extension phase during the stance phase of the subject's gait cycle; and obtaining an evaluation value to assess the condition of the knee by quantifying the variability of the helical axis group in one or both of the flexion phase and the extension phase.

Description

[Technical Field]

[0001] This invention relates to a gait analysis method and apparatus, and more specifically, to the analysis of knee joint movement of a subject during walking. [Background technology]

[0002] Osteoarthritis of the knee (KOA) is a common orthopedic disease affecting a large number of people worldwide. In KOA, joint deformation causes kinematic changes during walking, and walking function deteriorates significantly as the condition progresses. KOA is accompanied by cartilage wear, which is irreversible, and drugs and regenerative medicine that can regenerate cartilage have not yet been put into practical use. For this reason, early diagnosis and intervention are extremely important in KOA, and identifying biomarkers in the early stages of the disease is an urgent task.

[0003] Traditionally, knee thrust, a phenomenon characterized by lateral movement of the knee during walking, and an increase in knee varus moment (KAM) have been known to be risk factors for the progression of knee osteopathy (KOA). Methods have been proposed to diagnose KOA by detecting these factors using accelerometers and motion analysis devices. However, the occurrence of knee thrust and an increase in KAM indicates that KOA has progressed to a certain extent, and these indicators do not have sufficient detection power in the early stages, thus not contributing to the early diagnosis and treatment intervention of KOA.

[0004] The helical axis is a known method for representing the joint movement of the knee joint (see Patent Document 1 and Non-Patent Document 1). The helical axis is sometimes referred to as a spiral axis, and Patent Document 1 uses the term spiral axis. The relative movement between the femur and tibia can be described by rotation around a certain axis and translation along that axis, and that axis is the helical axis. The present inventors considered detecting early osteoarthritis of the knee by gait analysis focusing on the helical axis representing the joint movement of the knee.

[0005] Patent documents 2 to 6 are examples of patent documents related to the determination and detection of KOA (Knee Arthroplasty) based on the motion analysis of subjects during walking. Patent document 2 uses two acceleration sensors attached to the proximal tibia and proximal heel of the subject. Patent document 3 uses a three-axis acceleration sensor and a three-axis angular velocity sensor attached to the lower limb of the subject. Patent document 4 uses two IMUs attached to the thigh and lower leg of the subject. Patent document 5 evaluates the risk of knee injury of a subject based on the acceleration data of the subject during walking. Patent document 6 uses time-series data of foot position and knee position included in the subject's walking data, and this data is acquired by motion capture or inertial sensors. The motion analysis in patent documents 2 to 6 does not involve the use of a helical axis. It should be noted that while Non-Patent Literature 1 describes calculating the helical axis of the knee joint, the helical axis is calculated for every 10-degree change in knee movement. Considering that the range of motion of the knee during normal walking is approximately 10 degrees during the stance phase and 60 degrees during the swing phase, it is likely that the frequency of helical axis calculation is low. [Patent Document 1] Patent No. 6563418 [Patent Document 2] Patent No. 4350394 [Patent Document 3] Patent No. 6660110 [Patent Document 4] Patent No. 7541777 [Patent Document 5] Japanese Patent Publication No. 2017-202236 [Patent Document 6] Japanese Patent Publication No. 2023-167029 [Non-Patent Document 1] Dispersion of knee helical axes during walking in young and elderly healthy subjects, Federico Temporiti, et al, Journal of Biomechanics Volume 109, 26 August 2020 [Disclosure of the Invention] [Problems that the invention aims to solve]

[0006] The present invention aims to perform gait analysis that can be used to detect early-stage osteoarthritis of the knee by using the helical axis in the knee joint movement of a subject during walking. [Means for solving the problem]

[0007] The gait analysis method employed in this invention is: A process for calculating the helical axis of the knee joint based on the lower limb movement data of the subject during walking, The process of extracting the flexion phase and extension phase during the stance phase of the subject's gait cycle, A step of obtaining an evaluation value for evaluating the condition of the knee by quantifying the variation of the helical axis group in one or both of the flexion phase and the extension phase, Includes.

[0008] Standard deviation, mean deviation, and variance are examples of statistics used to quantify the variability of the helical axis, but the system is not limited to these, and other statistics may be used.

[0009] In one mode, the calculation frequency of the helical axis is higher than 10 Hz. In one mode, the calculation frequency of the helical axis is higher than 20 Hz. In one mode, the calculation frequency of the helical axis is higher than 30 Hz. In one mode, the calculation frequency of the helical axis is higher than 40 Hz. In one mode, the calculation frequency of the helical axis is higher than 50 Hz. In the embodiment described later, the calculation frequency of the helical axis is 200 Hz, but this value is not limited to that. Assuming that one walking cycle is about 1 second, when calculating the helical axis at 10 Hz, it will be about 10. If the stance phase is 0.6 seconds, the number of helical axes during the stance phase will be 6. When the calculation frequency of the helical axis is lower than 10 Hz, the number of helical axes obtained during the stance phase of one walking cycle will be reduced (even less in the flexion and extension phases), so there is a risk that the accuracy of quantifying the variation of the helical axis will be reduced. Note that the upper limit of the calculation frequency of the helical axis is not limited. However, in terms of the calculation frequency, the helical axis cannot be calculated above the sampling frequency. Also, by setting a threshold (defined by the knee joint angle) as a condition for calculating the helical axis, it should be noted that the upper limit of the number of helical axes obtained is limited.

[0010] In one aspect, the helical axis group is represented in spherical coordinates, and the standard deviation of one or both of the angle θ^ and the angle φ^ is used as the evaluation value. Note that the variation of the helical axis may also be quantified by the coordinates of the endpoints instead of the angles θ^ and φ^.

[0011] In one aspect, the helical axis group is approximated to a plane, and the average deviation of each helical axis with respect to the approximate plane is used as the evaluation value.

[0012] In one aspect, it includes a step of evaluating the risk of knee disorder of the subject based on the evaluation value. In a specific example of the evaluation step, the risk of knee disorder of the subject is evaluated by comparing the evaluation value of the subject with a reference value (set based on the evaluation value of a healthy person).

[0013] In one aspect, the helical axis is calculated using the first measurement value of the first inertial sensor attached to the thigh of the subject and the second measurement value of the second inertial sensor attached to the lower leg. The motion measurement unit for acquiring motion data for calculating the helical axis is not limited. However, the method using an inertial sensor is simpler, for example, compared to optical motion capture, and the burden on the subject is less. Also, by appropriately selecting the body part to attach the inertial sensor, more accurate data can be measured. An inertial sensor, i.e., an IMU, is a sensor equipped with a three-axis acceleration sensor and a three-axis angular velocity sensor.

[0014] The walking analysis device adopted in the present invention includes a helical axis calculation means for calculating the helical axis of the knee joint based on the movement data of the lower limbs of the subject during walking, a cutting-out means for cutting out the flexion phase and the extension phase during the stance phase of the walking cycle of the subject, and an evaluation value acquisition means for quantifying the variation of the helical axis group in one or both of the flexion phase and the extension phase to obtain an evaluation value for evaluating the state of the knee. It includes.

[0015] In one aspect, the helical axis calculation means calculates the helical axis at a frequency higher than 10 Hz. In one aspect, the helical axis calculation means calculates the helical axis at a frequency higher than 20 Hz. In one aspect, the helical axis calculation means calculates the helical axis at a frequency higher than 30 Hz. In one aspect, the helical axis calculation means calculates the helical axis at a frequency higher than 40 Hz. In one aspect, the helical axis calculation means calculates the helical axis at a frequency higher than 50 Hz.

[0016] In one aspect, the evaluation value acquisition means represents the helical axis group in spherical coordinate notation and obtains the standard deviation of one or both of the angle θ^ and the angle φ^ as the evaluation value.

[0017] In one aspect, the evaluation value acquisition means approximates the helical axis group to a plane, and obtains the average deviation of each helical axis with respect to the approximate plane as the evaluation value.

[0018] In one aspect, it includes an evaluation means for evaluating the risk of knee disorder of the subject based on the evaluation value.

[0019] The walking analysis device adopted in the present invention It consists of a computer equipped with a processor, The processor receives motion data of the subject's lower limbs during walking. The helical axis of the knee joint is calculated using the received motion data. Using the received motion data, the flexion and extension phases are extracted from the stance phase of the gait cycle. The system is configured to obtain an evaluation value for assessing the condition of the knee by quantifying the variation of the helical axis group in one or both of the flexion phase and the extension phase. The present invention is also provided as a computer program for executing the above-mentioned processor. In one embodiment, the computer is equipped with memory, which stores received motion data, calculated helical axes, and calculated evaluation values. [Effects of the Invention]

[0020] According to the present invention, by quantifying the movement of the knee joint during the stance phase based on the helical axis, an evaluation value that can be used to detect early osteoarthritis of the knee is obtained. [Brief explanation of the drawing]

[0021] [Figure 1] This is an overall diagram of the gait analysis system according to this embodiment. [Figure 2] This figure shows the hardware configuration of the gait analysis system according to this embodiment. [Figure 3] This figure shows the hardware configuration of the gait analysis system according to this embodiment. [Figure 4] This figure shows the calculation of the helical axis of the knee joint in the gait analysis system according to this embodiment. [Figure 5] This figure shows the analysis process in the gait analysis system according to this embodiment. [Figure 6] This figure shows the quantification of the helical axis variation in the analysis process according to this embodiment. [Figure 7] This figure shows the quantification of the helical axis variation in the analysis process according to this embodiment. [Figure 8] This figure shows the determination of KOA onset using the gait analysis system according to this embodiment. [Figure 9] This figure shows the means for acquiring lower limb movement data of a subject according to this embodiment. [Figure 10] This diagram illustrates the spherical coordinate representation of helical axes. [Figure 11] This diagram illustrates the planar approximation of the helical axis. [Figure 12] This shows the helical axis of a healthy individual (stance phase flexion). The diagrams show the helical axis in three planes (upper left: frontal plane, lower left: horizontal plane, upper right: sagittal plane), and the lower right diagram shows the axis from a direction that makes the changes in the axis easier to see. Figures 13 to 19 are similar, and in the lower right diagram, an approximate plane is also drawn when the variation is small. [Figure 13] This shows the helical axis of a healthy individual (during the stance phase extension). [Figure 14] This shows the helical axis of a healthy individual (swing phase, flexion phase). [Figure 15] This shows the helical axis of a healthy individual (swing phase extension). [Figure 16] This shows the helical axis of a patient with early-stage or mild KOA (during the stance phase flexion). [Figure 17] This shows the helical axis of a patient with early-stage or mild KOA (during the stance phase extension). [Figure 18] This shows the helical axis of a patient with early-stage or mild KOA (swing phase flexion). [Figure 19] This shows the helical axis of a patient with early-stage or mild KOA (swing phase extension). [Figure 20] This shows the evaluation (stance phase) using a spherical coordinate system, based on analysis of variance among three groups: young healthy individuals, middle-aged healthy individuals, and KOA patients (OA patients). [Figure 21] This shows the evaluation (swing phase) in a spherical coordinate system based on analysis of variance among three groups: young healthy individuals, middle-aged healthy individuals, and KOA patients (OA patients). [Figure 22]The evaluation using a planar approximation is shown based on analysis of variance for three groups: young healthy individuals, middle-aged healthy individuals, and KOA patients (OA patients). The upper figure represents the stance phase, and the lower figure represents the swing phase. [Modes for carrying out the invention]

[0022] [A] Overall configuration of gait analysis method and apparatus As shown in Figure 1, the gait analysis method according to this embodiment comprises the steps of: acquiring motion data of a subject during walking; calculating the helical axis of the knee joint based on the motion data; extracting the stance flexion phase, stance extension phase, swing flexion phase, and swing extension phase from the gait cycle based on the motion data (at least the stance flexion phase and stance extension phase need to be extracted); and obtaining evaluation values ​​to evaluate the state of the knee by quantifying the variability of the helical axis group in each phase (at least the stance flexion phase and stance extension phase). The motion data of a subject during walking is data from when the subject walks on the floor or on a treadmill. The motion data only needs to be able to calculate the helical axis of the knee joint and to allow extraction of at least the stance flexion phase and stance extension phase from the gait cycle; it does not necessarily need to be whole-body motion data, but it is sufficient if it includes at least lower limb motion data.

[0023] As shown in Figures 2 and 3, the gait analysis device according to this embodiment comprises a motion measurement unit that acquires motion data of a subject during walking, a helical axis calculation unit, and an analysis unit. The helical axis calculation unit and the analysis unit are composed of one or more computers. The computers include an input unit, a processor or processing unit, a memory or storage unit (RAM, ROM), and an output unit. In one embodiment, the output unit may include a display. The helical axis calculation unit calculates the helical axis based on the motion data measured by the motion measurement unit and a helical axis calculation program stored in memory, and the analysis unit performs gait analysis based on the calculated helical axis and a helical axis variation quantification program stored in memory, and acquires and outputs evaluation values ​​for evaluating the condition of the knee.

[0024] The motion measurement unit acquires motion data from at least the lower limbs (at least the thigh and lower leg) of the subject. In one embodiment, the motion measurement unit includes a first inertial sensor (IMU) located on the subject's thigh and a second inertial sensor (IMU) located on the lower leg. The inertial sensor typically comprises a 3-axis accelerometer, a 3-axis angular velocity sensor (gyroscope), and a geomagnetic sensor, although the geomagnetic sensor is an optional element. The first inertial sensor acquires 3-axis acceleration data and 3-axis angular velocity data of the thigh (femur) as first measurement data, and the second inertial sensor acquires 3-axis acceleration data and 3-axis angular velocity data of the lower leg (tibia) as second measurement data. The first measurement data from the first inertial sensor and the second measurement data from the second inertial sensor are synchronized and input from the computer's input unit to the helical axis calculation unit. In the configuration shown in Figure 4, the helical axis of the knee joint is calculated from the posture of the femur detected based on the first measurement data and the posture of the tibia detected based on the second measurement data.

[0025] The motion measurement unit may be a motion capture system. In one embodiment, the motion measurement unit is an optical motion capture system, comprising optical markers (feature points) placed at predetermined locations on at least the lower limbs (at least the thighs and lower legs) of the subject, and a plurality of synchronized video cameras that capture the movement of at least the lower limbs of the subject during walking. Time-series data of the 2D position information of each marker (feature point) is acquired and input from the computer's input unit, and the 3D position information of each marker (feature point) is reconstructed. The type of motion capture used in this embodiment is not limited, and examples include so-called markerless motion capture, which does not use optical markers or sensors. Examples of markerless motion capture include motion capture using a system equipped with a camera and a depth sensor (represented by Kinect), or motion capture that uses deep learning to analyze RGB images from one or more viewpoints to acquire motion data.

[0026] The relative movement between the femur and tibia can be described by rotation around a certain axis and translation along that axis, which is the helical axis. The helical axis can be calculated if time-series data of the femoral posture and time-series data of the tibia's posture are available, by representing these time-series data in a common coordinate system. It should be noted that the calculation of the helical axis does not necessarily require that the femoral posture data and tibia's posture data be output from the system. In other words, a helical axis calculation program is provided that takes motion measurement data (e.g., 3-axis acceleration data and 3-axis angular velocity data, or 2D position information of a marker) as input, calculates and outputs the helical axis, and it is also acceptable for only the helical axis to be calculated using predetermined inputs and output from the system.

[0027] A gait cycle can be defined as the period from when one foot touches the ground until the other foot on the same side touches the ground again, and is divided into the stance phase, which is the period when one foot is on the ground, and the swing phase, which is when that foot is off the ground. There are two types of knee movement during the stance and swing phases: flexion and extension. Flexion is the direction in which the foot moves towards the buttocks, while extension is the direction in which the foot moves away from the buttocks. If one of the flexion and extension directions is defined as the direction in which the knee joint angle decreases, then the other can be defined as the direction in which the knee joint angle increases. In other words, the stance and swing phases have a flexion phase and an extension phase, and the flexion phase of the stance phase, the extension phase of the stance phase, the flexion phase of the swing phase, and the extension phase of the swing phase can be extracted from the gait cycle, for example, based on time-series data of knee joint angles obtained from motion measurement data.

[0028] The helical axis calculation unit provides time-series data of the helical axis of the knee joint. The frequency of helical axis calculation can affect the accuracy of quantifying the variability of the helical axis group. If the number of helical axes obtained during the stance phase of one gait cycle decreases (and decreases even further during the flexion and extension phases), the accuracy of quantifying the variability of the helical axis may decrease. Furthermore, in order to detect and evaluate subtle differences in knee behavior, it is desirable to calculate the helical axis at appropriately small movements (angles). In one embodiment, the frequency of helical axis calculation is higher than 10 Hz. In one embodiment, the frequency of helical axis calculation is higher than 20 Hz. In one embodiment, the frequency of helical axis calculation is higher than 30 Hz. In one embodiment, the frequency of helical axis calculation is higher than 40 Hz. In one embodiment, the frequency of helical axis calculation is higher than 50 Hz. In the embodiment described later, the calculation frequency of the helical axis is 200 Hz, but it is not limited to this value. For example, any value between 50 Hz and 200 Hz may be used, or a frequency higher than 200 Hz may be used.

[0029] The flexion phase, extension phase, flexion phase, and extension phase of the swing phase, which are separated in the gait cycle, are identified by time, and the time-series data of the helical axis of the knee joint can be grouped into helical axis group 1 corresponding to the flexion phase of the stance phase, helical axis group 2 corresponding to the extension phase of the stance phase, helical axis group 3 corresponding to the flexion phase of the swing phase, and helical axis group 4 corresponding to the extension phase of the swing phase.

[0030] The number of gait cycles used to form each helical axis group is not limited; one gait cycle or multiple gait cycles may be used. For example, when forming helical axis group 1 based on two gait cycles, helical axis group 1 is formed from two helical axis groups: the helical axis group corresponding to the stance phase flexion phase of the first gait cycle and the helical axis group corresponding to the stance phase flexion phase of the second gait cycle. Furthermore, if the frequency of helical axis calculation is relatively low, the number of gait cycles used to form each helical axis group may be increased.

[0031] In the gait analysis unit, the variability of the helical axes in each of the helical axis groups, helical axis group 1, helical axis group 2, helical axis group 3, and helical axis group 4, is quantified. As will be described later, comparative experiments between KOA patients and healthy individuals have revealed differences in the patterns of axis variability during the stance phase. Therefore, in the gait analysis method and apparatus according to this embodiment, acquiring helical axis groups 3 and 4, and quantifying the variability of the helical axes in helical axis groups 3 and 4, is optional. That is, although Figures 5 to 7 show the quantification of the variability of the helical axes in helical axis group 3 and helical axis group 4, these are optional.

[0032] The statistical measures used to quantify the variability of the helical axes are not limited, and standard deviation, mean deviation, and variance are given as examples. In one embodiment, the variability of the helical axes is quantified by representing the group of helical axes in spherical coordinates and calculating the standard deviation of one or both of the angles θ^ and φ^. Note that it is sufficient to obtain the angles θ^ and φ^, and it is not necessary to actually display the spherical coordinates.

[0033] In one approach, the helical axis group is approximated by a plane, and the variation of the helical axis is quantified by calculating the average deviation between the plane and each axis. Note that actually displaying the approximated plane is not mandatory.

[0034] As shown in Figure 8, the analysis unit according to this embodiment may include a KOA onset detection unit. The KOA onset detection unit evaluates the risk of knee injury in a subject by comparing the evaluation value obtained in the subject with a reference value stored in memory (database). The reference value is typically a value set based on the evaluation value of a healthy person. If there are multiple types of statistics for quantifying variability, a reference value is prepared corresponding to each statistic. Multiple reference values ​​may be prepared according to the attributes of the subject. For example, multiple reference values ​​may be prepared for each age group.

[0035] There are no limitations on how the subject's evaluation values ​​and reference values ​​are used to determine the subject's risk of knee injury. For example, the risk may be determined by the degree of distance between the subject's evaluation value and the reference value, and the determination result may be output in two stages (e.g., "no abnormality," "risk present") or three stages. Furthermore, there are no limitations on how these evaluation values ​​are used to determine the risk of knee injury when multiple evaluation values ​​are obtained for a subject (e.g., when both the standard deviation and mean deviation are obtained). The determination result may be output for each evaluation value, or the risk of knee injury may be determined by comprehensively using multiple evaluation values. Alternatively, the risk of knee injury may be determined by combining the evaluation values ​​obtained in this embodiment with other evaluation values ​​or indicators different from those obtained in this embodiment. In addition, a reference table corresponding to the evaluation values ​​may be prepared, and the knee condition or risk of knee injury may be determined by comparing the output subject's evaluation values ​​with the reference table. It should be noted that evaluation values ​​according to this embodiment may be obtained for subjects wearing orthotics or insoles (e.g., KOA patients).

[0036] [B] Example [B-1] Overview As shown in Figure 9, a first inertial sensor is attached to the lateral side of the subject's thigh, slightly above the lateral femoral condyle, and a second inertial sensor is attached to the tibial tuberosity. Wireless communication is used to acquire first measurement data from the first inertial sensor and second measurement data from the second inertial sensor into a computer at 200 Hz. The inertial sensor is equipped with a 3-axis accelerometer, a 3-axis angular velocity sensor, and a geomagnetic sensor, but in this embodiment, the geomagnetic sensor is not used.

[0037] The subject is asked to walk approximately 6 meters from a stationary standing position. Based on the 3-axis acceleration and 3-axis angular velocity information obtained at this time, the following calculations are performed. (1) Synchronize the two inertial sensors. (2) Apply a low-pass filter with a cutoff frequency of 6 Hz. (3) Calculate the average value of the angular velocity when the person is standing still, and subtract that value from the angular velocity data. (4) The average value of the acceleration while standing at rest is calculated, and the direction of gravity is detected. Based on this, the coordinate system for standing at rest is defined. (5) The change in the attitude of the inertial sensor from a stationary standing position is calculated by integrating the angular velocity. (6) Determine the time change of the helical angle (i.e., the knee joint angle) between the two sensors. Note that the knee joint angle may also be the Euler angle or similar. (7) Using this, the flexion phase of the stance phase, the extension phase of the stance phase, the flexion phase of the swing phase, and the extension phase of the swing phase are extracted. (8) In the coordinate system of the inertial sensor attached to the thigh, determine the helical axis of the rotational motion of the inertial sensor on the lower leg. (9) The variation of the helical axis is represented in spherical coordinates. (10) The variation of the helical axis is represented by a planar approximation. A more detailed explanation follows.

[0038] [B-2] Calculation of IMU sensor attitude Perform zero-point correction for angular velocity. The upper left subscript S in TIFF2026081405000002.tif7157 indicates that it is described in the sensor coordinate system (Tib or Fem). S ω represents the 3-axis angular velocity of each sensor (femoral IMU, lower leg IMU). S ω0 is the average value of the three-axis angular velocity of each sensor (femoral IMU, lower leg IMU) when the body is standing still.

[0039] Global coordinate system as seen from the lower leg IMU coordinate system at rest (t=0) Tib0 M G This can be written as follows: TIFF2026081405000003.tif28156 Global coordinate system as seen from the IMU coordinate system of the thigh at rest Fem0 M G This can be written as follows: TIFF2026081405000004.tif30156a0 is the average value of acceleration in a stationary standing position. j and k are the coordinate axes (sensitivity axes) of the IMU. It is TIFF2026081405000005.tif35164.

[0040] The lower leg and thigh IMU coordinate systems at rest as seen from the Global coordinate system can be described as TIFF2026081405000006.tif7157.

[0041] The change in the sensor's attitude from rest is represented by the yxz Euler angles φ, θ, ψ (the initial values of φ, θ, ψ are zero). The rotation matrix representing the change in the sensor's attitude at that time can be described as follows. TIFF2026081405000007.tif49156

[0042] The rotation matrix S0 R S representing the attitude of each sensor at time t, and the angular velocity s ω (using the rotational speed around each axis, i.e., the derivative value) of the three axes of each sensor, the rotation matrix S0 R s representing the attitude of each sensor after Δt is obtained as follows. TIFF2026081405000008.tif47156

[0043] The lower leg and thigh IMU coordinate systems as seen from the Global coordinate system after Δt are described as follows. TIFF2026081405000009.tif7155

[0044] [B-3] Calculation of the helical axis Since the thigh IMU is attached such that the sensor z-axis faces the side of the body and the lower leg IMU faces the front of the body, basically, there is a 90-degree misalignment in orientation. Therefore, the thigh IMU coordinate system is rotated by 90 degrees to align the orientation with the lower leg IMU coordinate system. For the right thigh, it becomes TIFF2026081405000010.tif14164, and for the left thigh, it becomes TIFF2026081405000011.tif14164.

[0045] Determine the knee joint angle α. TIFF2026081405000012.tif14155 Here, H is the rotation matrix between the femoral IMU coordinate system and the lower leg IMU coordinate system.

[0046] The helical axis of the lower leg's postural changes is described in the femoral IMU coordinate system. TIFF2026081405000013.tif48155n is a unit vector on the helical axis. However, the helical axis calculation is performed only when the change in knee joint angle κ / Δt is greater than the threshold. Here, Δt = 0.05, and the thresholds used in the example are 30 deg / s for the stance phase and 120 deg / s for the swing phase. The helical axis at time t is determined between times A = t - Δt / 2 and B = t + Δt / 2. If Δt is 0.05 and the change in knee joint angle during this period does not meet the threshold, the helical axis is not calculated. If it is greater than the threshold, the helical axis is calculated at 200 Hz.

[0047] The threshold value used in calculating the helical axis can be appropriately set by those skilled in the art. Since the axis becomes indeterminate if there is no change in the angle of the knee joint, in order to properly calculate the helical axis, which is the axis of motion, it is necessary for the knee joint to move to some extent, i.e., to set a large threshold value. On the other hand, if the threshold value is set too large, the helical axis may not be properly calculated during the stance phase, when the change in the knee joint angle is limited. The current threshold value is 30 deg / s during the stance phase, but it is not limited to this value, and a smaller threshold value may be used, for example.

[0048] [B-4] Quantification of helical axis variation The knee joint angle α typically flexes and extends twice during one walking cycle (double knee action). The extreme values ​​are determined, and the following phases are extracted: (1) flexion phase during the stance phase, (2) extension phase during the stance phase, (3) flexion phase during the swing phase, and (4) extension phase during the swing phase. The variability in each of these phases is quantified using the following two methods.

[0049] [B-4-1] Spherical coordinate representation of helical axis group unit vector of the helical axis obtained The TIFF2026081405000014.tif10146 file is expressed in spherical coordinates as follows (see Figure 10). TIFF2026081405000015.tif29146

[0050] As shown in Figure 6, the standard deviations of θ^ and φ^ in the spherical coordinate representation of the helical axis group are calculated for each phase (flexion phase of the stance phase, extension phase of the stance phase, flexion phase of the swing phase, and extension phase of the swing phase) to quantify the variability.

[0051] [B-4-2] Axis deviation from the plane approximating the motion of the helical axis When the least squares approximation plane of the helical axis group in each phase is ax+by+cz=d, the deviation ε from the plane to the i-th axis is... i teeth The filename becomes TIFF2026081405000016.tif17155 (see Figure 11).

[0052] As shown in Figure 7, the deviation ε between each helical axis and the approximate plane is as follows: i We calculate the average value and quantify the variability.

[0053] [C] Experimental example [C-1] Experimental conditions Using the method described in the above example, the experiment was conducted in the following procedure. Based on the lower limb motion data of the subject during walking (first measurement data from the first inertial sensor, second measurement data from the second inertial sensor), the helical axis of the knee joint is calculated. In the gait cycle of the subjects, the flexion phase during the stance phase, the extension phase during the stance phase, the flexion phase during the swing phase, and the extension phase during the swing phase were extracted. The variation in the helical axis group in each phase was quantified. The helical axis was acquired for one gait cycle.

[0054] The subject's profile is as follows: Five young healthy individuals (M3, F2), average age 29.0 years. Nine middle-aged healthy individuals (M4, F5), average age 49.3 years. KOA patients (OA): 9 (M2, F7), average age 68.6 years The age shown is the average age for each group. Please note that the KOA patient group consists of patients with relatively early-stage and mild cases.

[0055] [C-2] Display of helical axis Figures 12-15 show the helical axis of a healthy individual (stance phase flexion), a healthy individual (stance phase extension), a healthy individual (swing phase flexion), and a healthy individual (swing phase extension). The healthy individuals here are young individuals. In Figures 12-15, the helical axis is viewed in three planes (upper left: frontal plane, lower left: horizontal plane, upper right: sagittal plane), and the lower right is a view from a direction that makes the change in the axis easier to see. In the lower right, an approximate plane is also drawn when the variation is small.

[0056] Figures 16-19 show the helical axis of a KOA patient (stance phase flexion), a KOA patient (stance phase extension), a KOA patient (swing phase flexion), and a KOA patient (swing phase extension). In Figures 16-19, the helical axis is viewed in three planes (upper left: frontal plane, lower left: horizontal plane, upper right: sagittal plane), and the lower right is viewed from a direction that makes the change in the axis easier to see. In the lower right, an approximate plane is also drawn when the variation is small.

[0057] [C-3] Analysis of the helical axis For the statistical analysis, we performed an analysis of variance (ANOVA) on three groups based on the obtained helical axis: young healthy individuals, middle-aged healthy individuals, and KOA patients. [C-3-1] Analysis 1 In each phase, namely the flexion phase of the stance phase, the extension phase of the stance phase, the flexion phase of the swing phase, and the extension phase of the swing phase, the helical axis groups obtained for each subject were represented in spherical coordinates, and the standard deviations of θ^ and φ^ were calculated for each phase to quantify the variability of the helical axes. Figure 20 shows the evaluation using spherical coordinates (stance phase). A significant difference was found between healthy individuals and KOA patients in the standard deviation of θ^ (extension phase). In Figure 21, no significant difference was found between healthy individuals and KOA patients in the evaluation using spherical coordinates (swing phase).

[0058] [C-3-2] Analysis 2 In each phase, namely the flexion phase of the stance phase, the extension phase of the stance phase, the flexion phase of the swing phase, and the extension phase of the swing phase, the helical axis groups obtained for each subject were approximated in a plane, and the mean deviation of each helical axis relative to the approximation plane was calculated to quantify the variability of the helical axes. Figure 22 shows the evaluation using the plane approximation, with the upper figure representing the stance phase and the lower figure representing the swing phase. In the stance phase, significant differences were found between healthy individuals and KOA patients in both the mean deviation (flexion phase) and the mean deviation (extension phase). In the swing phase, no significant differences were found between healthy individuals and KOA patients.

[0059] [C-4] Findings It was found that there is a difference in the pattern of helical axis variability during the stance phase between KOA and healthy individuals. In other words, it is considered effective to perform gait analysis to obtain evaluation values ​​for the knee condition by calculating the helical axis of the knee joint based on the lower limb movement data of the subject during walking, extracting the flexion phase and extension phase during the stance phase of the subject's gait cycle, and quantifying the variability of the helical axis group in one or both of the flexion phase and extension phase. Furthermore, the middle-aged healthy group is a kind of pre-KOA group and may include early KOA patients, but a difference in helical axis variability was observed between the middle-aged healthy group and the KOA patient group. This suggests that the evaluation values ​​for assessing the knee condition according to this embodiment can be used to detect early osteoarthritis of the knee.

Claims

1. A process for calculating the helical axis of the knee joint based on the lower limb movement data of the subject during walking, The process of extracting the flexion phase and extension phase during the stance phase of the subject's gait cycle, A step of obtaining an evaluation value for evaluating the condition of the knee by quantifying the variation of the helical axis group in one or both of the flexion phase and the extension phase, A gait analysis method that includes this.

2. The calculation frequency for the helical axis is higher than 10Hz. The gait analysis method according to claim 1.

3. The helical axis group is represented in spherical coordinates, angle angle The evaluation value is the standard deviation of one or both of the following: A gait analysis method according to either claim 1 or 2.

4. Approximate the helical axis group in a plane, The evaluation value is the average deviation of each helical axis relative to the approximation plane. A gait analysis method according to either claim 1 or 2.

5. The process includes evaluating the risk of knee injury to the subject based on the aforementioned evaluation values. A gait analysis method according to either claim 1 or 2.

6. The helical axis is calculated using the first measurement value from the first inertial sensor attached to the subject's thigh and the second measurement value from the second inertial sensor attached to the lower leg. A gait analysis method according to either claim 1 or 2.

7. A helical axis calculation means that calculates the helical axis of the knee joint based on the movement data of the subject's lower limbs during walking, A means for extracting the flexion and extension phases during the stance phase of a subject's gait cycle, An evaluation value acquisition means for obtaining an evaluation value for evaluating the condition of the knee by quantifying the variation of the helical axis group in one or both of the flexion phase and the extension phase, A gait analysis device that includes a gait analysis system.

8. The helical axis calculation means calculates the helical axis at a frequency higher than 10 Hz. The gait analysis device according to claim 7.

9. The aforementioned evaluation value acquisition means expresses the helical axis group in spherical coordinates, angle angle Obtain the standard deviation of one or both of the following as the evaluation value. A gait analysis device according to any one of claims 7 or 8.

10. The means for obtaining the evaluation value is, Approximate the helical axis group in a plane, The average deviation of each helical axis relative to the approximation plane is obtained as the evaluation value. A gait analysis device according to any one of claims 7 or 8.

11. Includes an evaluation means for evaluating the risk of knee injury in a subject based on the aforementioned evaluation value, A gait analysis device according to any one of claims 7 or 8.

12. The helical axis calculation means calculates the helical axis using the first measurement value of the first inertial sensor attached to the thigh of the subject and the second measurement value of the second inertial sensor attached to the lower leg. A gait analysis device according to any one of claims 7 or 8.

13. It consists of a computer equipped with a processor, The processor receives motion data of the subject's lower limbs during walking. The helical axis of the knee joint is calculated using the received motion data. Using the received motion data, the flexion and extension phases are extracted from the stance phase of the gait cycle. The system is configured to obtain an evaluation value for assessing the condition of the knee by quantifying the variation of the helical axis group in one or both of the flexion phase and the extension phase. Walking analysis device.

14. A computer program for causing the processor described in claim 13 to run.

Images