Motion analysis device, motion analysis method, and program

Abstract

A motion analysis device (1) is provided with: measurement units (21, 31) for measuring, as myoelectric data and kinematic data, the movement of a joint and an end point that are moved by an antagonistic muscle group; an augmented synergy matrix calculation unit (14) for calculating a muscle synergy matrix described by a radial direction vector and a deflection-angle direction vector, the muscle synergy matrix associating a displacement vector of a muscle antagonistic ratio obtained from the myoelectric data with a displacement vector of an equilibrium point position of the end point obtained from the myoelectric data and the kinematic data; and an equilibrium point manipulability calculation unit (15) for calculating, as analysis information, an equilibrium point manipulability from the muscle synergy matrix. This device enables a muscle synergy for evaluating a motion state of a human body to be analyzed using the equilibrium point manipulability as an indicator.

Description

Motion analysis device, motion analysis method, and program

[0001] The present invention relates to a motion analysis technique for analyzing muscle synergies to evaluate the motion status of a human body using an index called equilibrium point manipulability.

[0002] Various methods and devices have been proposed for analyzing muscle group states based on muscle synergies obtained by measuring the activities of multiple muscles during exercise (e.g., Patent Documents 1 and 2). Patent Document 1 describes a motion analysis device that obtains muscle synergies using the muscle-antagonism ratio of electromyographic data during exercise to evaluate sports exercise in an engineering-based quantitative evaluation. Patent Document 2 describes a motion analysis device that regards physical movement as a mechanical system of musculoskeletal structures and introduces the concepts of muscle-antagonism ratio and muscle-antagonism sum to calculate analytical information such as a stiffness ellipse, an equilibrium point, and muscle synergies.

[0003] Furthermore, Non-Patent Document 1 mathematically shows, from a kinematic viewpoint, how freely the position and posture of a robot arm can be manipulated in a workspace when the end effector is attached to the tip of the arm in the field of robotics, and describes an analytical method for quantitatively evaluating the degree of freedom based on the size of a geometric feature such as the volume of a manipulability ellipsoid.

[0004] On the other hand, in medical terms, proximal femoral fractures are common among the elderly in Japan, and rehabilitation aimed at restoring walking is widely implemented (Non-Patent Document 2).On the other hand, in rehabilitation for neurological disorders such as stroke, attention is focused on global physical function (coordinated activity of muscle groups), given that walking is a whole-body movement (Non-Patent Document 3).

[0005] Patent No. 5158824 Patent No. 6518932

[0006] Tsuneo Yoshikawa, Fundamentals of Robot Control, pp. 109-131, Corona Publishing, 1988. Edited by the Japanese Orthopaedic Association / Japan Society for Fracture Therapy, and edited by the Japanese Orthopaedic Association Clinical Practice Guidelines Committee / Femoral Neck / Trochanteric Fracture Clinical Practice Guidelines Development Committee: Femoral Neck / Trochanteric Fracture Clinical Practice Guidelines 2021 [Revised 3rd Edition], Nanzando, 2021. D. J. Clark, L. H. Ting, F. E. Zajac, R. R. Neptune, and S. A. Kautz: “Merging of healthy motor modules predicts reduced locomotor performance and muscle coordination complexity post-stroke,” J Neurophysiol, vol. 103, no. 2, pp. 844-857, 2010. N. J. Fairhall, S. M. Dyer, J. C. Mak, J. Diong, W. S. Kwok, and C. Sherrington: “Interventions for improving mobility after hip fracture surgery in adults,” Cochrane Database Syst Rev, vol. 9, no. 9, CD001704, 2022.

[0007] Recently, various proposals for more accurately evaluating and analyzing the motion state of the human body have been desired in rehabilitation aimed at functional recovery from musculoskeletal disorders and training aimed at improving sports skills. For example, the analytical method described in Non-Patent Document 1 (2013) is considered to be applicable to the human body. However, Non-Patent Document 1 describes a control technology applied to robotics, which typically utilizes a mapping matrix (Jacobian matrix) between the displacement of the joint variable vector of a robot manipulator and the displacement of the position / posture vector of the end-effector at the tip of the manipulator. Therefore, it is not easy to directly apply this manipulability concept to a mapping matrix for human motion analysis focusing on muscle synergies, considering that human joint motion involves redundant multiple muscle groups and that human motion is significantly affected by not only kinematic aspects such as joint angles but also dynamic aspects such as mechanical impedance.

[0008] Furthermore, according to Non-Patent Document 4, rehabilitation for proximal femoral fractures has traditionally tended to focus on local physical functions such as muscle strength at the injured site, and there has been limited evidence regarding gait improvement. Furthermore, the global perspective of physical function in gait reconstruction in Non-Patent Document 3 is also important for diseases that do not directly affect the central nervous system, and detailed research is considered necessary into the relationship between proximal femoral fractures and muscle coordination.

[0009] The present invention has been made in view of the above, and provides a motion analysis device, a method and a program that enable analysis of muscle synergies for evaluating the motion status of a human body by applying an index called equilibrium point manipulability.

[0010] The motion analysis device according to the present invention includes: a measuring unit which measures movements of a joint and an end point, which are moved by antagonistic muscle groups, as electromyographic data and kinematic data; a muscle synergy matrix calculating means which calculates a muscle synergy matrix described by radius vectors and deflection angle direction vectors, which correlates a displacement vector of a muscle antagonist ratio obtained from the electromyographic data with a displacement vector of an equilibrium point position of the end point obtained from the electromyographic data and the kinematic data; and an analytical information calculating means which calculates an equilibrium point manipulability as analytical information from the muscle synergy matrix.

[0011] Furthermore, the motion analysis method according to the present invention includes the steps of: causing a computer to measure, by a measuring unit, movements of a joint and an end point, which are moved by antagonistic muscle groups, as electromyographic data and kinematic data; calculating a muscle synergy matrix described by radial and angular vectors, which correlates a displacement vector of a muscle antagonistic ratio obtained from the electromyographic data with a displacement vector of an equilibrium point position of the end point obtained from the electromyographic data and the kinematic data; and calculating, from the muscle synergy matrix, equilibrium point manipulability as analysis information.

[0012] The program according to the present invention causes a computer to function as: muscle synergy matrix calculating means for calculating a muscle synergy matrix described by radius vectors and deflection angle direction vectors, which correlates a displacement vector of a muscle antagonistic ratio obtained from the myoelectric data with a displacement vector of an equilibrium point position of the endpoint obtained from the myoelectric data and the kinematic data, among the myoelectric data and kinematic data measured by a measuring unit, which reflect the movements of a joint and an endpoint moved by antagonistic muscles; and analysis information calculating means for calculating equilibrium point manipulability as analysis information from the muscle synergy matrix.

[0013] According to these inventions, the measurement unit measures the movements of the joints and end points moved by antagonistic muscle groups as electromyographic data and kinematic data, the muscle synergy matrix calculation means calculates a muscle synergy matrix described by radius vectors and deflection angle direction vectors that correlate a displacement vector of the muscle antagonist ratio obtained from the electromyographic data with a displacement vector of the equilibrium point position of the end point obtained from the electromyographic data and the kinematic data, and the analysis information calculation means calculates equilibrium point manipulability as analysis information from the muscle synergy matrix. Therefore, muscle synergies for evaluating the motion status of a human body can be analyzed by applying an index called equilibrium point manipulability. Note that equilibrium point manipulability represents the ease of manipulating the equilibrium point at a body end point, and can be used as an index for measuring the level of functional recovery or skill improvement in motor skills, for example.

[0014] According to the present invention, muscle synergies for evaluating the motion status of a human body can be analyzed by applying an index called equilibrium point manipulability.

[0015] 1 is a diagram explaining manipulability described in Non-Patent Document 1, where (A) shows the state of change in the shape of the manipulability ellipsoid, and (B) shows the characteristics of manipulability.

[0023] Fig. 1 is a diagram approximating the lower limbs moving in the sagittal plane using a link model, where (A) explains the muscle groups around the femur, and (B) explains the kinematic coordinates.

[0024] Fig. 1 is a circuit diagram showing an embodiment of a motion analysis device according to the present invention.

[0025] Fig. 2 is a flowchart explaining an embodiment of the motion analysis processing procedure.

[0026] Fig. 1 is a diagram showing the equilibrium point manipulability for each subject before and after rehabilitation intervention.

[0027] Fig. 2 is a diagram showing the relationship between the degree of assistance and the equilibrium point manipulability.

[0028] Fig. 1 is a conceptual diagram of a configuration for measuring the walking pattern of a subject.

[0029] Fig. 2 corresponds to Fig. 2 and is applied in Verification Experiment 2, where (A) explains the muscle groups around the femur, and (B) explains the kinematic coordinates.

[0029] Fig. 1 is a triangular diagram showing the coactivation synergy of three cases that were typical in terms of walking ability at the time of discharge, where (A) is before rehabilitation (at the start of walking with a cane) and (B) is after rehabilitation (at the time of discharge). Figure 1 shows the equilibrium point manipulability of each subject before and after rehabilitation. Figure 2 shows box plots of the equilibrium point manipulability of each independent and non-independent group before and after rehabilitation. Figure 3 shows the change over time of the equilibrium point manipulability ellipse (dotted line) during one gait cycle at the start of walking with a cane for (A) the fractured side and (B) the non-fractured side. Figure 4 shows the change over time of the equilibrium point manipulability ellipse during one gait cycle at the time of discharge from hospital for (A) the fractured side and (B) the non-fractured side.

[0016] First, the underlying technology and outline of the present invention will be described.

[0017] (1) Manipulability in Robotics (Base Technology) Equation 1 represents formulas (A) to (F) that explain manipulability in robotics.

[0018]

[0019] To explain equations (A) to (C) in Math 1, the upper right corner shows a mechanical diagram of a two-link model robot arm in which a base arm and a hand arm are connected. In this mechanical diagram, when the first joint is θ1 and the second joint is θ2, the position and movement amount of the end point of the hand arm are expressed as in equations (A) to (C). Furthermore, as can be seen from the matrix expression in equation (D), when the Jacobian matrix J(θ) described in terms of joint angles for mapping transformation is applied to the displacement Δθ of each joint angle, the displacement Δx of the motion position of the tip is shown.

[0020] In Equation (C), the displacements Δθ1 and Δθ2 of the two joint angles exist within a unit circle in the joint space (joint displacement space), and the positional displacements Δx and Δy of the end points at that time are located within an ellipse in the projected workspace (work displacement space). The ellipse is described based on Equation (E). The larger the area of ​​the ellipse, the larger the positional displacements Δx and Δy, making it possible to operate within a larger ellipse, i.e., improving manipulability. In robotics control, the area of ​​the ellipsoid is calculated by applying Equation (F) to the matrix J(θ), and this is used as the manipulability w, an index for evaluating and analyzing ease of operation. Furthermore, according to Equation (E), the shape of the ellipse in the workspace changes depending on the posture of each arm, revealing changes in manipulability in terms of the directions in which the hand is easy and difficult to operate. Figure 1 shows the relationship between the shape of the manipulability ellipsoid and each posture of a two-link arm in which each link is the same length, with the area of ​​the ellipsoid changing according to the posture of the arm, i.e., the joint angle displacements Δθ1 and Δθ2, and as shown in Figure 1(B) in particular, the area is maximum (maximum manipulability) when the angle between the two links is 90 degrees, and decreases from there in both directions. In other words, because the area of ​​the manipulability ellipsoid increases and decreases according to the characteristics of formula (F), knowing the area of ​​the ellipsoid makes it possible to recognize the current degree of ease of operability (i.e., it can serve as an index).

[0021] (2) Equilibrium point manipulability in human muscle groups Equilibrium point manipulability in human body movement is evaluated from the dynamics point of view, that is, from the viewpoint of the impedance balance of muscle groups, regarding the manipulability of the equilibrium points of the end points.

[0022] (2.1) Lower Limb Model Figure 2 shows an approximation of the lower limb moving in the sagittal plane using a link model. (A) illustrates the muscle groups around the femur, and (B) illustrates the kinematic coordinates. This model has six major lower limb muscles: the gluteus maximus (GM), iliopsoas (IL), semitendinosus (ST), rectus femoris (RF), vastus lateralis (VL), and short head of the biceps femoris (BF), forming antagonistic muscle pairs around the hip joint, hip and knee joint, and knee joint (Table 1). Table 1 shows the muscle-antagonist ratio, muscle-antagonist sum, and their functions.

[0023]

[0024] The end point (ankle joint position) is expressed by polar coordinates (R, Φ) centered on the hip joint (FIG. 2(B)).

[0025] (2.2) Co-activation synergy Co-activation synergy is a hypothetical concept of muscle coordination, which introduces two variables, the muscle antagonist ratio and the muscle antagonist sum, to calculate the equilibrium point of the endpoint in this model (Fig. 2). Its validity has been verified theoretically and experimentally as shown in the following references [1] to [5]. ([1]: F. Miyazaki, H. Hirai, M. Uemura, K. Uno, and T. Oku: "Motion analysis apparatus, method for analyzing motion, and motion analysis program," US10631751B2, 2020. [2]: Kenta Tominaga, T. Iimura, M. Uemura, H. Hirai, and Fumio Miyazaki: "Muscle synergy as an invariant related to human lower limb movement," Transactions of the Society of Instrument and Control Engineers, vol. 52, no. 1, pp. 37-45, 2016. [3]: E. Watanabe, K. Kozasa, R. Fujihara, H. Hirai, K. Yoshida, H. Naritomi, and H. I. Krebs: “Exploiting the invariant structure for controlling multiple muscles in anthropomorphic legs: III. Reproducing hemiparetic walking from equilibrium point-based synergies,” in Proc 2019 IEEE 16 thInt Conf Rehabil Rob (ICORR2019), pp. 1227-1232, 2019. [4]:E. Watanabe, H. Hirai, and H. I. Krebs: “Equilibrium point-based control of muscle-driven anthropomorphic legs reveals modularity of human motor control during pedaling,” Adv Rob, vol. 34, no. 5, pp. 328-342, 2020. [5]:K. Noro, A. Takeichi, H. Hirai, H. Okamoto, D. Kogawa, C. Kamimukai, Y. Kaneko, N. Tabuchi, S. Kinoshita, Y. Miyajima, Y. Yashima, H. Kuga, S. Yamamoto, N. Yamada, K. Matsui, A. Nishikawa, and H. I. Krebs: “Physiological markers of motor improvement following five-month sprint training in young boys,” in Proc 11 th Int Symp Adaptive Motion of Animals and Machines (AMAM2023), pp. 124-125, 2023. ).

[0026] In addition, the aforementioned literature [1] indicates that the muscle antagonist ratio contributes to the equilibrium joint angle, and the muscle antagonist sum is involved in joint impedance. j and muscle antagonists j are defined as the following equations 2.

[0027]

[0028] where r jindicates the muscle antagonist ratio at joint j (j = h[hip], hk[hip and knee], k[knee]), the denominator is the sum of the muscle activities of the extensor and flexor muscles of the antagonistic muscle pair, and the numerator is the muscle activity of the extensor muscle (Equation (1)). j is the sum of muscle antagonists for joint j, which represents the total muscle activity of the extensor and flexor muscles of the antagonistic muscle pair (Equation (2)).

[0029] In the lower limb model shown in Figure 2, the equation for the balance of forces between the hip joint and knee joint is arranged using the variables of muscle antagonist ratio and muscle antagonist sum, and the displacement of the equilibrium joint angle of the hip joint and knee joint (Δθ h,EP , Δθ k,EP ) T is expressed as follows, as shown in Equation 3:

[0030]

[0031] In equation (3) of Mathematical Expression 3, A is a constant determined by the moment arm of the joint and the natural length of the muscle. Furthermore, equation (3) calculates a vector consisting of the displacement of the equilibrium joint angle by mapping a vector consisting of the displacement of the muscle antagonistic ratio information with a matrix consisting of the constant A and the muscle antagonistic sum information of equations (4) and (5).

[0032] On the other hand, the displacement of the equilibrium point position of the end point (ΔR, ΔΦ) T is the equilibrium joint angle displacement (Δθ h,EP , Δθ k,EP ) T Using the above, it can be expressed as in equation (6): where L is the link length of the thigh and the lower leg (the same length).

[0033]

[0034] Here, when equation (3) is substituted into equation (6), equations (7) and (8) are obtained.

[0035]

[0036] Here, B(θ k , EP ) is a value determined by the joint moment arm, natural muscle length, link length, and equilibrium joint angle of the knee joint, and can be approximated by a constant for a certain movement. EP , ΔΦ EP ]T , s=[s h ,s hk ,s k ] T , Δr = [Δr h , Δr hk , Δr k ] T where S(s) represents the first two matrices on the right side of equation (7). Equation (7) expresses the displacement vector [Δr h , Δr hk , Δr k ] T This means that the displacement of the foot equilibrium point can be estimated by projecting and scaling it. Since q2 is a vector in the radial direction and q1-q2 / 2 is a vector in the argument direction, by finding the basis vectors in each direction, we obtain equations (9) and (10).

[0037]

[0038] u R , u Φ is the displacement vector of the muscle antagonistic ratio [Δr h , Δr hk , Δr k ] T These quantities characterize the distribution of the radial and angular directions of the u, which are defined as the radial and angular co-active synergy vectors, respectively. Furthermore, as additional information, u, which contributes to the adjustment of the impedance of the endpoints, R×Φ = (u R ×u Φ ) / │u R ×u Φ | is defined as the coactivation synergy vector in the null space direction. Here, we use the term coactivation synergy because these vectors are variables determined only by the sum of muscle antagonisms and represent the impedance balance of muscles.

[0039] (2.3) Equilibrium Point Manipulability First, in robot control, the concept of manipulability is widely known as a method for evaluating the manipulability of an end effector attached to the end of an arm from a kinematic point of view (Non-Patent Document 1). When the joint variable vector of the robot arm is θ and the tip position vector is x, the manipulability w is defined as shown in Equation (12) using the Jacobian matrix J(θ) that appears in Equation (11) which is the relation between the joint angle displacement vector Δθ and the tip position displacement vector Δx, which are minute changes between them.

[0040]

[0041]

[0042] Correspondingly, the equilibrium point manipulability w ＥＰ When the expanded synergy matrix S appearing in equations (8) and (equation (H) in equation 10) is defined as shown in equation (13), the manipulability of the equilibrium point w ＥＰ is a quantity that reflects the area of ​​the ellipsoid corresponding to the manipulability of the end point with respect to the equilibrium point. Therefore, by analyzing the increase in the area of ​​the equilibrium point manipulability ellipsoid estimated from the extended synergy matrix S that reflects the body movement, it is possible to confirm the ease of body movement control.

[0043]

[0044] In addition, in the equation (7), the displacement vector Δr of the muscle antagonism ratio h , Δr hk , Δr k exists within the unit sphere of the muscle antagonistic ratio space (muscle antagonistic ratio displacement space), and the displacement ΔR of the equilibrium point position of the end point at that time is EP , ΔΦ EP is located within an ellipse in the work space (work displacement space) expressed in the polar coordinate system, and this relationship is shown in equation (G) in equation 10. Here, S ＋ represents the pseudo-inverse matrix of the extended synergy matrix S. Therefore, if the area of ​​the ellipse is large, the displacement ΔR of the equilibrium point position of the end point EP , ΔΦ EP becomes larger, allowing for operation over a wider range, i.e., the smooth point manipulability improves.

[0045] Regarding the manipulability of the end points, as shown in equation (8) (equation (H) in equation 10), the equilibrium point manipulability w ＥＰ can be a new index for evaluation from the dynamic viewpoint, that is, from the viewpoint of the impedance balance of the muscle group, as shown in equation (13).

[0046]

[0047] Next, FIG. 3 is a circuit diagram showing one embodiment of a motion analysis device according to the present invention. The motion analysis device 1 is used to evaluate and analyze the motion state of a human body in rehabilitation aimed at functional recovery from musculoskeletal disorders or training aimed at improving sports skills. The motion analysis device 1 includes an electromyographic signal detection circuit 21 that measures electromyographic signals of muscle groups, such as antagonistic muscles around joints, generated during a predetermined human motion, and an imaging unit 31 as a motion measurement unit that measures the movements of the endpoints of the motion location and the joints. The detected signals are input to a signal processing unit 10 and subjected to predetermined signal processing. The signal processing unit 10 is typically configured as a computer with a built-in processor.

[0048] An operation unit 101, a storage unit 102, and a display unit 41 are connected to the signal processing unit 10. The operation unit 101 is used to input instructions and necessary information from outside. The storage unit 102 includes a RAM and a ROM; the RAM temporarily stores measurement data and data in the middle of processing, and the ROM stores signal processing programs and various data required for processing. The display unit 41 not only allows the user to confirm the content input via the operation unit 101, but also displays the results of processing by the signal processing unit 10 as image information in a predetermined format. The operation unit 101 may be configured as a touch panel superimposed on the screen of the display unit 41.

[0049] The signal processing unit 10 functions as an electromyography data processing unit 11, a kinematics data processing unit 12, a normalization processing unit 13, an extended synergy matrix calculation unit 14, an equilibrium point manipulability calculation unit 15, and a display processing unit 16 by reading and executing the signal processing program stored in the memory unit 102.

[0050] The myoelectric signal detection circuit 21 has a required number of myoelectric electrodes 211 attached individually to the skin surface at the center of each of multiple antagonistic muscles to be measured during exercise. The myoelectric signal detection circuit 21 or the myoelectric electrodes 211 may be wired or wireless myoelectric sensors (e.g., PicoEMG, Cometa S.r.l., Italy). The myoelectric signal detection circuit 21 synchronously detects myoelectric potentials from a predetermined number of myoelectric electrodes 211, transmits the detected myoelectric data from each muscle to the signal processing unit 10, and stores it in the memory unit 102. In a mode in which myoelectric signals are measured during abnormal gait by attaching electrodes to six muscle groups around the femur (proximal femur) of the lower limb, it is preferable to attach electrodes to both legs (a total of 12 electrodes on both the affected and healthy sides) and acquire each myoelectric signal from both legs in a distinguishable manner. The myoelectric signal detection circuit 21 acquires myoelectric signals at a predetermined detection period. The myoelectric data processing unit 11 filters the myoelectric data taken into the memory unit 102 using a band-pass filter (e.g., a fourth-order Butterworth filter, 20-450 Hz), rectifies it, and removes outliers, then normalizes it to the myoelectric potential at the time of maximum voluntary contraction (MVC) and takes it in again.

[0051] The imaging unit 31 may be a video camera (e.g., HDR-CX 680, SONY Corp., Japan), which captures images from the sagittal plane at a predetermined frequency, e.g., 30 Hz, so as to orient the image of the athlete in a predetermined direction. For example, in the case of walking, the captured image (kinematic data) is input to the signal processing unit 10 via a wired or wireless connection and stored in the memory unit 102. The kinematic data processing unit 12 reads the kinematic data stored in the memory unit 102, digitizes it using, for example, open-source motion analysis software Kinovea, and re-records it as kinematic data for target joints, e.g., the hip joint, knee joint, ankle joint, heel, and fifth metatarsal. The digitizing process refers to, for example, discretely sampling the position and movement (displacement) information of target joints, e.g., the hip joint, knee joint, ankle joint, heel, and fifth metatarsal, from the captured human body image. The device for measuring kinematic data is not limited to an imaging device.

[0052] The normalization processing unit 13 synchronizes the recording of the electromyographic data and the kinematic data in the time axis direction, i.e., normalizes them. For example, if the kinematic data is walking movement, one walking cycle is defined as the time from heel strike to the same side heel strike again, and the electromyographic data and the kinematic data are normalized in the time axis direction.

[0053] The extended synergy matrix calculation unit 14 analyzes the measurement data consisting of the electromyographic data and the kinematic data normalized by the normalization processing unit 13 by applying a synergy analyzer or the like, estimates co-activation synergy vectors from the analyzed measurement data, and calculates the extended synergy matrix S(s) by using the estimated co-activation synergy vectors. The synergy analyzer is an analysis technology corresponding to Patent Document 2, and executes the calculation of Equation (7).

[0054] More specifically, the extended synergy matrix calculation unit 14 executes equation (2) from the EMG data, followed by equations (4) and (5), and then applies the kinematic data to calculate the product of the first two matrices on the right side of equation (7), i.e., the extended synergy matrix S(s). Equation (7) is a vector of displacement [Δr h , Δr hk , Δr k ] T This means that the displacement of the foot tip balance point can be estimated by projecting and scaling it. Also, since q2 is a vector in the radial direction and q1-q2 / 2 is a vector in the angular direction, the basis vectors for each direction can be calculated from equations (9) and (10) as the co-active synergy vector u Ｒ , u Φ It is calculated as follows.

[0055] The equilibrium point manipulability calculation unit 15 calculates the equilibrium point manipulability w by applying equation (13) to the extended synergy matrix S(s) calculated by equation (7) or equation (8). ＥＰ The display processing unit 16 displays the numerical information calculated by the equilibrium point manipulability calculation unit 15, or displays an image in a predetermined format.

[0056] 4 is a flowchart showing the procedure for the motion analysis process. First, electromyographic electrodes are attached to a predetermined muscle of the body part to be measured, and the imaging area is set in a predetermined direction so as to include the body part, and measurement data acquisition begins (step S1). Measurement is performed to detect electromyographic data and kinematic data, which are the measurement data, and it is determined whether measurement has ended or not when a predetermined time or predetermined motion has been performed (step S3). Note that all or a required portion of the data acquired during the measurement operation may be used for analysis.

[0057] When it is determined that the measurement is completed, the normalization processing unit 13 synchronizes the recorded myoelectric data and kinematic data in the time axis direction, i.e., normalizes them, upon receiving an instruction for analysis processing subsequently or afterwards (step S5). Next, the extended synergy matrix calculation unit 14 analyzes the measurement data consisting of the normalized myoelectric data and kinematic data, estimates coactivation synergy vectors from the analyzed measurement data, and calculates the extended synergy matrix S(s) using the estimated coactivation synergy vectors (step S7).

[0058] Next, the equilibrium point manipulability calculation unit 15 applies equation (13) to the extended synergy matrix S(s) to calculate the equilibrium point manipulability w ＥＰ (Step S9). Then, the display processing unit 16 calculates the equilibrium point manipulability w calculated by the equilibrium point manipulability calculation unit 15. ＥＰ and other numerical information are displayed on the display unit 41 (step S11). The analysis results displayed on the display unit 41 may be stored in the storage unit 102 or a cloud server in association with ID information, and may be displayed alongside new analysis results from subsequent analyses of the same person for comparison. This allows changes such as recovery of motor function and improvement in skills to be confirmed.

[0059] [Verification Experiment 1] Next, we will explain the verification experiment 1 conducted to confirm the effectiveness. Verification Experiment 1 considered rehabilitation intervention and gait control of a patient after surgery for a proximal femur fracture from the perspectives of muscle coordination and equilibrium point manipulability, and was conducted using the following procedure.

[0060] (1) Subjects: Four elderly women admitted to a medical center in Osaka Prefecture after surgery for a proximal femoral fracture participated in the study. The purpose and content of the experiment were explained to the participants, and their consent was obtained. This experiment was conducted with the approval of the Osaka University Research Ethics Committee. The subjects were able to walk with a cane before injury, had sufficient cognitive function, and had no other conditions that affected their walking. Measurements were conducted both at the start of cane walking and at the time of discharge, while walking with a cane. Each subject exhibited a gradation in walking ability: one who used a wheelchair for mobility and only walked with a cane during rehabilitation (S1, S2), one who required supervision for cane walking (S3), and one who was stable with a cane during ward life and could walk independently with supervision (S4). Basic information and clinical evaluations are shown in Table 2.

[0061]

[0062] (2) Equipment and Procedure: Subjects were instructed to walk 10 m in a rehabilitation room, three times in both directions, for a total of six trials, with sufficient rest time between each trial. Electromyographic electrodes were attached to the relevant muscles of both legs (affected and unaffected), and muscle activity during walking was recorded with a wireless electromyographic sensor (PicoEMG, Cometa S.r.l., Italy) at a sampling frequency of 2000 Hz. Measurements were performed on 12 muscles in both legs, including six muscles: gluteus maximus (GM), iliopsoas (IL), semitendinosus (ST), rectus femoris (RF), vastus lateralis (VL), and short head of biceps femoris (BF). In parallel, a video camera (HDR-CX 680, SONY Corp., Japan) captured sagittal plane walking at a frame rate of 30 Hz. The video images were digitized using Kinovea, an open-source motion analysis software, to record kinematic information for the hip, knee, ankle, heel, and fifth metatarsal. Electromyogram and kinematic recordings were synchronized.

[0063] (3) The electromyographic data obtained by analysis was filtered with a band-pass filter (fourth-order Butterworth filter, 20-450 Hz), rectified, and outliers were removed. The data was then normalized by the electromyographic potential at maximum voluntary contraction (MVC). Furthermore, the period from heel strike to ipsilateral heel strike again was defined as one gait cycle, and the electromyographic data and kinematic data were normalized in the time direction. Measurement data for six gait cycles of both legs recorded from a 10-m walk (six trials) was analyzed using the synergy analyzer. After estimating coactivation synergy, the manipulability of the equilibrium point was calculated.

[0064] (4) Results (4.1) Walking Ability Table 3 shows the walking ability of the four subjects before and after rehabilitation intervention.

[0065]

[0066] The Timed Up and Go (TUG) test is a clinical evaluation that measures the time it takes to stand up from a chair, walk 3 meters back and forth, and sit down again, and is widely known and used as a fall predictor for the elderly. Assistance Levels 1 to 4 represent the walking aids used when moving around independently within a hospital, with each aid providing increasing amounts of physical assistance to support the patient's weight in the order of 1: walking alone, 2: cane, 3: walker, and 4: wheelchair, and the lower the level, the greater the walking ability.

[0067] (4.2) Balance Point Manipulability Figure 5 is a chart showing the balance point manipulability for each subject before and after rehabilitation intervention. The vertical axis represents the value of balance point manipulability, with white representing the value before rehabilitation (when walking with a cane begins) and gray representing the value after rehabilitation (when discharged from hospital). Subject numbers S1 to S4 are assigned to the horizontal axis, and the subjects are arranged from left to right in order of decreasing walking ability based on the clinical assessment before rehabilitation. Figure 6 is a chart showing the relationship between assistance levels 1 to 4 and balance point manipulability. The assistance levels are displayed in order of decreasing balance point manipulability. Note that Figure 6 shows a total of eight subjects, with the results for each subject taken twice, before and after rehabilitation.

[0068] (5) Discussion: Before and after rehabilitation, subjects S1 and S2 (see Table 3), who required less physical assistance while walking, showed an increase in equilibrium point manipulability (see Figure 5), indicating that equilibrium point manipulability responds to changes in the amount of physical assistance while walking before and after rehabilitation. Furthermore, the numerical values ​​for equilibrium point manipulability generally increased after rehabilitation, allowing quantitative recognition of the improvement in motor control of the endpoints associated with the recovery of motor function in the target body parts.

[0069] Furthermore, regardless of the time of measurement, a relationship (correlation) has been observed between the amount of physical assistance during walking and equilibrium point manipulability, for example, the lower the equilibrium point manipulability, the higher the amount of physical assistance (see Figure 6). Therefore, equilibrium point manipulability not only reflects changes in the amount of physical assistance required for walking, but also has the potential to serve as an index for determining the amount of physical assistance. In other words, instead of determining the amount of physical assistance based on subjective evaluation, it is possible to objectively estimate the amount of physical assistance based on equilibrium point manipulability. For example, equilibrium point manipulability can be divided into four categories corresponding to the amount of physical assistance, and the amount of physical assistance corresponding to the category of the calculated equilibrium point manipulability value can be presented.

[0070] A survey of patients with proximal femoral fractures who were living in the community after rehabilitation showed that those who achieved cane walking at the time of discharge also had high walking ability six months after discharge. Reducing the amount of physical assistance required during hospitalization is one of the major rehabilitation goals. Traditionally, performance indicators such as the TUG (Training Guide) have been used to determine whether a patient can walk independently in the hospital, and gait pattern has been assessed based on the fall risk cutoff value. However, this study shows that the TUG does not necessarily correlate with the amount of physical assistance required for walking (see Table 3). Clinically, we often encounter cases where walking speed has improved but unsteadiness is observed, making it difficult to walk with a cane in the hospital. In such cases, qualitative assessment, such as visual inspection, has been necessary. However, because equilibrium point maneuverability can sensitively capture the amount of physical assistance required during walking, which is difficult to determine based on walking speed alone, it can provide a consistent understanding of cases where the amount of physical assistance required is difficult to determine using conventional methods, and may serve as a more useful guide for selecting walking aids.

[0071] Furthermore, as a result of the analysis, an equilibrium point manipulability ellipse may be calculated and displayed on the display unit 41. In this case, the ease of manipulating the equilibrium point manipulability ellipse is on the major axis side, and when the flattening is high, the major axis side is easier to manipulate.

[0072] Furthermore, the co-activation synergy vector in the null space direction may be added to the expanded synergy matrix S(s) applied to this process. That is, when the relationship between the three muscle antagonistic ratios and each displacement of the equilibrium point position is expressed in two directions, that is, in the radial R direction and the deflection angle Φ direction, the null space N direction component of R×Φ is further added to express it as a 3×3 square matrix (Equation (I) in Equation 11). As a result, an ellipsoid can be drawn by Equation (J) in Equation 11, and the equilibrium point manipulability reflecting the volume of the ellipsoid can be calculated by Equation (13).

[0073]

[0074] Furthermore, although this embodiment has mainly described the analysis of motor function recovery around the hip joint, femur, and knee joint of the lower limbs and their antagonistic muscles, the present invention is not limited to this and can be similarly applied to any moving part having a skeleton of two upper limb links consisting of the upper arm and forearm, and two lower limb links consisting of the thigh and lower leg.Furthermore, the present invention can be applied to evaluating and analyzing the state of recovery in rehabilitation after surgery for injury or fracture, as well as evaluating and analyzing the progress and guidance of motor skills in sports.Furthermore, in this invention, a cane is not required for walking during tests for motion analysis, and any walking aid other than a cane can be used as long as appropriate measurement data can be obtained.

[0075] As described above, Verification Experiment 1 suggested that there is an activity pattern of antagonistic muscle pairs in people who are able to walk independently with a cane upon discharge from hospital, and that gait assessment based on muscle coordination, known as equilibrium point manipulability, is useful.

[0076] Next, in Verification Experiment 2, evaluations were conducted on a larger number of subjects. As a result, new muscle activity patterns were observed in those who were able to walk with a cane upon discharge. At the same time, a more detailed analysis of motor control after a proximal femur fracture was conducted by evaluating the equilibrium point manipulability ellipse from a geometric perspective in addition to the degree of equilibrium point manipulability. The details of this analysis are explained in [Verification Experiment 2] below.

[0077] [Verification Experiment 2] (1) Overview of Verification Experiment 2 In Verification Experiment 2, we examined the gait recovery of 13 hip fracture patients using three motor control indices (coactivation synergy, balance point manipulability, and toe stiffness). As a result, we obtained two important findings, as described below. First, patients who achieved cane walking were classified into at least two groups based on coactivation synergy. This suggests that there are multiple solutions for gait recovery. Second, balance point manipulability is closely related to coactivation synergy and toe stiffness and may be associated with patients' independent mobility. This suggests that balance point manipulability may be a useful therapeutic indicator. These findings may be useful for the development of future exercise interventions.

[0078] (2) Method (2.1) Subjects The subjects were 13 elderly people admitted to Hoshigaoka Medical Center after surgery for proximal femoral fractures. The purpose and content of the experiment were explained to them, and their consent to participate was obtained. The experiment was conducted with the approval of the Osaka University Research Ethics Committee. The subjects were required to be able to walk with a cane before their injury, to have sufficient cognitive function, and to have no other illnesses that could affect their walking. Measurements were taken at the start of cane walking and at the time of discharge. Table 4 shows basic information about the subjects at the start of cane walking, divided according to whether they were able to walk independently with a cane at the time of discharge.

[0079]

[0080] (2.2) Measurements: As in [Verification Experiment 1], subjects walked 10 m round trips three times, for a total of six trials. Muscle activity in both legs (affected and unaffected) during walking was recorded at 2000 Hz using a wireless electromyography (EMG) sensor (PicoEMG, Cometa Srl, Italy). Six muscles were measured: the gluteus maximus (GM), iliopsoas (IL), semitendinosus (ST), rectus femoris (RF), vastus lateralis (VL), and short head of the biceps femoris (BF). In parallel, sagittal plane walking was filmed at 30 Hz using a video camera (HDR-CX680, SONY® Corp, Japan) (Figure 7). The footage was digitized using motion analysis software Kinovea, and kinematic information was recorded for the hip, knee, ankle, heel, and fifth metatarsal. The electromyography and kinematic recordings were synchronized. Clinical assessments included pain, fall risk measure (References [6], [7]), and assistance level.Pain was assessed using the walking pain item of the Western Ontario and McMaster Universities (WOMAC) Osteoarthritis Index ([6]: N. Bellamy, W. W. Buchanan, C. H. Goldsmith: “Validation study of WOMAC: A health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee,” J Rheumatology, vol. 12, pp. 1833-1840, 1988.), and fall risk was assessed using the Time Up and Go Test ([7]: A. Shumway-Cook, S. Brauer, and M. Woollacott: “Predicting the probability for falls in community-dwelling older adults using the Timed Up & Go Test,” Physical Therapy, vol. 80, no. 9, pp. 896-903, 2000.). The degree of walking assistance was calculated by quantifying the walking style when moving around alone according to the amount of assistance needed (1: walking alone, 2: cane, 3: walker, 4: wheelchair) and used as an index of walking ability.

[0081] (2.3) Analysis: The acquired EMG data was band-pass filtered (fourth-order Butterworth filter, 20-450 Hz), rectified, and outlier-removed before being normalized by the EMG during maximum voluntary contraction (MVC). The data was then normalized in the time direction, with one gait cycle defined as the time from heel strike to ipsilateral heel re-striking. The data was analyzed using a synergy analyzer (see reference [2]) to estimate coactivation synergies, which were then combined with the kinematic data to calculate movement features representing the equilibrium point manipulability. The calculation methods are briefly described below. Further details are provided in [Verification Experiment 1].

[0082] (2.3.1) Co-activation synergy Co-activation synergy is a hypothetical concept of muscle coordination that introduces two variables, the muscle antagonist ratio and the muscle antagonist sum, into the motor control of the endpoints of a musculoskeletal model, and its validity has been verified through theory and experiments (see references [1] and [2]). The muscle antagonist ratio is related to the equilibrium joint angle, and the muscle antagonist sum is related to the joint impedance (see reference [1]), and they are defined as shown in Table 5.

[0083]

[0084] When the movement of the end point (ankle joint) of the model (Figure 8) consisting of the six main muscles is expressed in polar coordinates (R, Φ) centered on the hip joint, the change in its equilibrium position (ΔR ＥＰ , ΔΦ ＥＰ ) Ｔ can be expressed as equation (14) using the muscle antagonist ratio, muscle antagonist sum, and equilibrium joint angle.

[0085]

[0086] where q 1、 q2 is calculated from the following equations (16) and (17).

[0087]

[0088] A is a constant determined by the moment arm of the joint and the natural length of the muscle, and B(θ ｋ，ＥＰ ) is a value determined by the joint moment arm, muscle natural length, link length, and equilibrium joint angle of the knee. Also, ΔP=[ΔR ＥＰ , ΔΦ ＥＰ ] Ｔ , s = [s h， s hk， s k ] T , Δr = [Δr h， Δr hk， Δr k ] T In equation (15), S(s) represents the product of the first two matrices on the right-hand side of equation (14) (extended synergy matrix). Equation (14) means that the displacement vector Δr of the muscle antagonistic ratio can be projected onto the space spanned by vectors q2 and q1-q2 / 2, and the displacement of the equilibrium point of the toe can be estimated. These represent vectors in the radial and angular directions, respectively, and their basis vectors u R , u Φis a quantity that characterizes the distribution of the muscle antagonism ratio in the radius and angular directions of the displacement vector Δr, and is calculated using equations (18) and (19).

[0089]

[0090] These are defined as co-active synergy vectors in the radial and angular directions, and are used to define the vectors that do not contribute to the displacement of the foot balance point. NULL = (u R ×u Φ ) / |u R ×u Φ Define | as the coactivation synergy vector in the null space.

[0091] The relationship between each synergy will now be explained. First, a coactivation synergy or a coactivation synergy matrix is ​​a combination of radial and angular coactivation synergy vectors, and the coactivation synergy matrix plus kinematic information is the extended synergy matrix (Equation (14) and Equation (15)). The extended synergy matrix is ​​equal to the muscle synergy matrix, and the muscle synergy matrix is ​​equal to the muscle synergy.

[0092] (2.3.2) Equilibrium point manipulability In robotics, the concept of manipulability (Non-Patent Document 1), which is widely known as an index for evaluating the kinematic manipulation ability of an arm end point, has been expanded and a new definition of equilibrium point manipulability has been made this time as an index for evaluating the dynamic manipulation ability of a musculoskeletal limb end point. Of these, the equilibrium point manipulability degree w ＥＰ is calculated and numerically evaluated using the expanded synergy matrix S appearing in equation (15) in equation (20).

[0093]

[0094] The equilibrium manipulability ellipse is the expanded synergy matrix S and its transpose matrix S Ｔ The ellipsoid is calculated from the eigenvalues ​​and eigenvectors of the product and evaluated from a geometrical aspect.

[0095] (2.3.3) Toe stiffness ellipse When an external force acts on the toes from the equilibrium state, the toe stiffness matrix K that relates the restoring torque generated in the hip and knee joints to the toe position. ｘ is calculated using equation (21) and can be geometrically expressed as an ellipse (see reference [2]).

[0096]

[0097] Here, C is a constant determined by the proportionality constant relating muscle activity and muscle stiffness and the moment arm of each joint, and J is the Jacobian matrix ∂(x, y) / ∂(θ h ,θ k )

[0098] (2.4) Statistical Processing: For clinical evaluation and balance point manipulability, patients were divided into an independent group and an independent group based on whether they used a cane at the time of discharge, and comparisons between groups were made using the Mann-Whitney U test. Furthermore, comparisons before and after rehabilitation were made using the Wilcoxon signed-rank test. The significance level was set at 5%.

[0099] (3) Results (3.1) Clinical Evaluations Table 6 shows the clinical evaluations of the independent and non-independent groups at the time of starting walking with a cane and at the time of discharge (before / after rehabilitation).

[0100]

[0101] (3.2) Coactivation synergy Figure 9 shows a triangular diagram of the coactivation synergy of three patients who were typical in terms of walking ability at the time of discharge. (A) shows before rehabilitation (when walking with a cane started) and (B) shows after rehabilitation (at the time of discharge). The radial coactivation synergy vector u Ｒ is shown in white, and the co-active synergy vector u Φ is shown in gray, and the co-active synergy vector u ＮＵＬＬ are shown in black. The triangles represent the fractured side, the circles represent the non-fractured side, and the stars represent the average of both sides. The three sides of the triangle represent the proportion of antagonistic muscles in the co-activation synergy vector. The graphs are arranged as follows: (A) on the left represents before rehabilitation (at the start of walking with a cane), (B) on the right represents after rehabilitation (at the time of discharge), the top row represents a representative of the non-independent group (S1), and the middle and bottom rows represent two representatives of the independent group (S5, S10).

[0102] (3.3) Equilibrium point manipulability (3.3.1) Equilibrium point manipulability Figure 10 shows the equilibrium point manipulability for each subject. The vertical axis represents equilibrium point manipulability, and the horizontal axis represents the subject label. Gray represents before rehabilitation (when walking with a cane begins), and black represents after rehabilitation (when discharged from hospital). Figure 11 also shows a box plot of the equilibrium point manipulability for the independent and non-independent groups at each time point. The vertical axis represents equilibrium point manipulability, and the box plot shows the median and interquartile range as boxes, with the upper and lower whiskers representing the maximum and minimum values.

[0103] (3.3.2) Equilibrium Point Manipulability Ellipse Figure 12 shows the change over time of the equilibrium point manipulability ellipse (dotted line with black circles and white numbers) during one gait cycle before rehabilitation (at the start of walking with a cane). Figure 13 shows the change over time of the equilibrium point manipulability ellipse (dotted line with black circles and white numbers) during one gait cycle after rehabilitation (at the time of discharge from hospital). The graphs in each figure are arranged such that the left side represents (A) the fractured side and the right side represents (B) the non-fractured side, and the top, middle, and bottom rows represent three people: a representative of the non-independent group (S1), a representative of the independent group (S5), and a representative of the independent group (S10).

[0104] (3.4) Toe Stiffness Ellipse Figure 12 shows the time change of the toe stiffness ellipse (gray solid line with white circles and black numbers) during one gait cycle before rehabilitation (at the start of walking with a cane). Figure 13 shows the time change of the toe stiffness ellipse (gray solid line with white circles and black numbers) during one gait cycle after rehabilitation (at the time of discharge from hospital). In this example, the toe stiffness ellipse represented by the gray solid line has a high degree of flattening, and appears almost like a line segment in the figure. The arrangement of the graphs in each figure is the same as that of the equilibrium point manipulability ellipse.

[0105] (4) Discussion (4.1) Coactivation Synergies as an Indicator of Independent Walking. Focusing on the arrangement of the average coactivation synergies of the left and right legs, indicated by stars in the triangular diagram in Figure 9, two patterns are evident in subjects who were able to walk independently with a cane at the time of discharge. These patterns are similar to those observed in subject S5 (middle row), where each synergy vector (white star, gray star, black star) is aligned along the dotted midline. These patterns are similar to those observed in subject S10 (bottom row), where each synergy vector spreads out to each vertex. At the muscle pair level, these patterns differ in the proportion of hip monoarticular muscle pairs and biarticular muscle pairs in the deviation direction. For example, subject S5 has an equal proportion of both muscle pairs, while subject S10 has a greater proportion of biarticular muscles. Subject S10 had a history of a contralateral fracture, and under these special circumstances, the contribution of biarticular muscle pairs to deviation movement may be greater. Moreover, the arrangement of subject S1 is roughly intermediate between those of subjects S5 and S10, and from these facts, it is possible that guiding the subject to either a linear arrangement (middle row) or a triangular vertex arrangement (lower row) is a condition for independent walking from the viewpoint of muscle coordination. For example, by displaying the triangular diagram shown in Fig. 9 on the display unit 41 via the display processing unit 16 and displaying the patient's radius vector, declination angle, and co-activation synergy vectors of the null space direction, or by calculating the coordinates of each co-activation synergy vector for the equilibrium point manipulability calculation unit 15, it becomes possible to support pattern recognition from their positions and their mutual positional relationships.

[0106] (4.2) Equilibrium Point Manipulability as an Indicator of Mobility Form. Observing the equilibrium point manipulability of each subject before and after rehabilitation in the bar graphs in Figure 10, we see that subjects with lower levels of walking assistance at each time point generally have higher values, reflecting changes before and after rehabilitation. However, several subjects experienced a decrease in equilibrium point manipulability after rehabilitation. These subjects corresponded to those whose level of walking assistance remained unchanged or decreased. In particular, subject S8's request to be discharged using a pushcart during rehabilitation may have been reflected in the training content. The results in Figures 10 and 11 suggest that the group who were able to walk independently with a cane after rehabilitation (at the time of discharge) had statistically significantly higher equilibrium point manipulability, which may serve as an indicator for assessing mobility forms at that time during rehabilitation. However, there was significant variability in equilibrium point manipulability before rehabilitation (at the start of cane walking), which is likely due to variations in risk assessment among therapists, as this was the beginning of training. By using the equilibrium point manipulability, it is possible to make an objective assessment at that time, and if the appropriate form of movement can be determined early on, it is possible that daily activities themselves will become rehabilitation, increasing the possibility of regaining walking.

[0107] (4.3) Toe Stiffness Ellipse and Equilibrium Point Manipulability Ellipse Visualizing Walking Control Strategies The toe stiffness ellipses shown in Figures 12 and 13 are larger during the stance phase and smaller during the swing phase, regardless of the level of independence in cane walking. However, the independent group exhibits a greater variety of sizes. This suggests that the higher the walking ability, the more selectively the stiffness can be adjusted. Furthermore, the equilibrium point manipulability ellipses are larger in the independent cane walking group, and the direction of their major axes varies. This suggests that the higher the walking ability, the more margin there is for the direction of movement of the endpoints that can be selected at any given time. Furthermore, even when comparing subjects who were able to walk independently with a cane at the time of discharge, subject S5, whose coactivation synergy (Figure 9) was linear, had a rounded equilibrium point manipulability ellipse, while subject S10, whose coactivation synergy spread to the vertices of a triangle, had a less rounded equilibrium point manipulability ellipse. This suggests that even if subjects are able to walk independently, the strategically prioritized movement direction differs.

[0108] Thus, according to [Verification Experiment 2], patients with proximal femoral fractures were evaluated from the perspective of muscle coordination and motor control variables, and it was suggested that those who were ultimately able to walk independently with a cane had a certain muscle coordination pattern, and that it was useful to incorporate equilibrium point manipulability into clinical evaluations.

[0109] As described above, the motion analysis device according to the present invention includes a measuring unit which measures the movements of joints and end points moved by antagonistic muscle groups as electromyographic data and kinematic data, a muscle synergy matrix calculating means which calculates a muscle synergy matrix described by radius vectors and deflection angle direction vectors which correlate a displacement vector of a muscle antagonist ratio obtained from the electromyographic data with a displacement vector of an equilibrium point position of the end point obtained from the electromyographic data and the kinematic data, and an analytical information calculating means which calculates the equilibrium point manipulability as analytical information from the muscle synergy matrix.

[0110] Furthermore, the motion analysis method according to the present invention includes the steps of: causing a computer to measure, by a measuring unit, movements of a joint and an end point, which are moved by antagonistic muscle groups, as electromyographic data and kinematic data; calculating a muscle synergy matrix described by radial and angular vectors, which correlates a displacement vector of a muscle antagonistic ratio obtained from the electromyographic data with a displacement vector of an equilibrium point position of the end point obtained from the electromyographic data and the kinematic data; and calculating, from the muscle synergy matrix, equilibrium point manipulability as analysis information.

[0111] The program according to the present invention causes a computer to function as: muscle synergy matrix calculating means for calculating a muscle synergy matrix described by radius vectors and deflection angle direction vectors, which correlates a displacement vector of a muscle antagonistic ratio obtained from the myoelectric data with a displacement vector of an equilibrium point position of the endpoint obtained from the myoelectric data and the kinematic data, among the myoelectric data and kinematic data measured by a measuring unit, which reflect the movements of a joint and an endpoint moved by antagonistic muscles; and analysis information calculating means for calculating equilibrium point manipulability as analysis information from the muscle synergy matrix.

[0112] According to these inventions, the measurement unit measures the movements of the joints and end points moved by antagonistic muscle groups as electromyographic data and kinematic data, the muscle synergy matrix calculation means calculates a muscle synergy matrix described by radius vectors and deflection angle direction vectors that correlate a displacement vector of the muscle antagonist ratio obtained from the electromyographic data with a displacement vector of the equilibrium point position of the end point obtained from the electromyographic data and the kinematic data, and the analysis information calculation means calculates equilibrium point manipulability as analysis information from the muscle synergy matrix. Therefore, muscle synergies for evaluating the motion status of a human body can be analyzed by applying an index called equilibrium point manipulability. Note that equilibrium point manipulability represents the ease of manipulating the equilibrium point at a body end point, and can be used as an index for measuring the level of functional recovery or skill improvement in motor skills, for example.

[0113] Furthermore, when the muscle antagonistic ratio space (muscle antagonistic ratio displacement space) is within a unit sphere, the analysis information calculation means calculates, as the equilibrium point manipulability, a value reflecting the area of ​​an ellipse expressed in the working space (working displacement space) after transformation with the muscle synergy matrix. According to this configuration, the value reflecting the area of ​​the ellipse is analyzed as the equilibrium point manipulability.

[0114] Furthermore, the analysis information calculation means creates an ellipse having an area reflecting the equilibrium point manipulability in a working space (working displacement space) having two axes, the radius vector and the deflection angle direction, from the muscle synergy matrix. With this configuration, an ellipse having an area reflecting the equilibrium point manipulability is created, with the good and bad manipulability directions as the major and minor axes, respectively. The characteristics of the equilibrium point manipulability can be visually recognized from the shape of the ellipse.

[0115] Furthermore, when the joints are the hip joint and knee joint sandwiching the femur, the end point is the ankle joint, and the movement is walking, the analysis information calculation means calculates information on the amount of object assistance during walking according to the equilibrium point manipulability. According to this configuration, by utilizing the correlation between the equilibrium point manipulability during walking and the amount of object assistance, the amount of object assistance can be obtained once the equilibrium point manipulability is known based on measurement data.

[0116] Furthermore, the present invention is provided with a display for displaying an image, and the analysis information calculation means is provided with display processing means for displaying the calculation results on the display. With this configuration, the calculated equilibrium point manipulability and, if necessary, an elliptical figure (size, shape (orientation, oblateness)) are visually displayed on the display unit.

[0117] REFERENCE SIGNS LIST 1 Motion analysis device 21 Myoelectric signal detection circuit (measurement unit) 211 Myoelectric electrode (measurement unit) 31 Imaging unit (measurement unit) 10 Signal processing unit 11 Myoelectric data processing unit 12 Kinematic data processing unit 13 Normalization processing unit 14 Extended synergy matrix calculation unit (muscle synergy matrix calculation means) 15 Equilibrium point manipulability calculation unit (analysis information calculation means) 16 Display processing unit 102 Storage unit 41 Display unit

Claims

1. A motion analysis device comprising: a measurement unit that measures the movements of joints and end points moved by antagonistic muscle groups as electromyographic data and kinematic data; a muscle synergy matrix calculation means that calculates a muscle synergy matrix described by radius and deflection direction vectors that correlates a displacement vector of a muscle antagonist ratio obtained from said electromyographic data with a displacement vector of an equilibrium point position of said end point obtained from said electromyographic data and said kinematic data; and an analysis information calculation means that calculates equilibrium point manipulability from said muscle synergy matrix as analysis information.

2. The motion analysis device according to claim 1, wherein the analysis information calculation means calculates, as the equilibrium point manipulability, a value reflecting the area of ​​an ellipse expressed in the working space after transformation with the muscle synergy matrix, when the muscle antagonistic ratio space is within a unit sphere.

3. The motion analysis device according to claim 1, wherein the analysis information calculation means calculates an elliptical figure corresponding to the equilibrium point manipulability from the synergy matrix in a working space having the radius and declination directions as two axes.

4. A motion analysis device as described in claim 1, wherein when the joints are the hip joint and knee joint sandwiching the femur, the end point is the ankle joint, and the motion is walking, the analysis information calculation means calculates information on the amount of object assistance during walking according to the equilibrium point manipulability.

5. A motion analysis device according to any one of claims 1 to 3, further comprising a display for displaying an image, wherein the analysis information calculation means comprises display processing means for displaying the calculation results on the display.

6. A movement analysis method comprising the steps of: a computer causing a measuring unit to measure the movements of a joint and an end point, which are moved by antagonistic muscle groups, as electromyographic data and kinematic data; a computer calculating a muscle synergy matrix described by radial and angular vectors, which relates a displacement vector of a muscle antagonist ratio obtained from the electromyographic data to a displacement vector of an equilibrium point position of the end point obtained from the electromyographic data and the kinematic data; and a computer calculating an equilibrium point manipulability from the muscle synergy matrix as analysis information.

7. A muscle synergy matrix calculation means for calculating a muscle synergy matrix described by radial and angular vectors that correlates a displacement vector of the muscle antagonist ratio obtained from the electromyographic data and a displacement vector of the equilibrium point position of the endpoint obtained from the electromyographic data and the kinematic data, which are among the electromyographic data and kinematic data measured by the measuring unit and reflect the movements of the joints and endpoints that are moved by antagonistic muscles; and a program for causing a computer to function as an analytical information calculation means for calculating the equilibrium point manipulability from the muscle synergy matrix as analytical information.

Images