Walking function evaluation device and its control program

Abstract

To provide a walking function evaluation device and a walking function evaluation method capable of improving remarkably quick organization of a treatment planning of a subject by recognizing a secular change of a walking function of the subject.SOLUTION: A first signal pattern of a biopotential signal of a subject, and a second signal pattern of a biopotential signal equivalent to a healthy person are normalized and both are compared. Thus, a walking function of the subject is recognized as a secular change based on similarity obtained from a comparison result.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a walking function evaluation device and The control program In particular, a walking function evaluation device for patients with progressive neuromuscular diseases to improve their walking function by wearing a wearable movement assist device, and The control program This is what we are trying to propose. [Background technology]

[0002] Progressive neuromuscular diseases such as amyotrophic lateral sclerosis (ALS) and muscular dystrophy (MD) are caused by nerve or muscle damage, gradually resulting in muscle weakness and motor dysfunction. There are no curative treatments for these diseases, and drug treatments have been unable to do more than slow the natural progression of symptoms.

[0003] Various power assist devices have been widely used to assist or substitute for the movements of physically disabled people who have lost muscle strength, elderly people who have weakened muscle strength, etc. As one of these power assist devices, for example, a wearable motion assist device has been proposed that can control and assist movement based on the bioelectric potential associated with voluntary muscle activity according to the wearer's intention (see Patent Document 1).

[0004] In recent years, such wearable motion-assist devices have been used in treatments aimed at maintaining and improving the walking function of patients with progressive neuromuscular diseases. These wearable motion-assist devices move in unison with the patient to assist walking movements based on physiological and motor information such as bioelectrical signals (BES) of the lower limb muscles, joint angles, and ground reaction forces.

[0005] A subject wearing this wearable motion-assist device can repeatedly walk based on the patient's motor intentions without placing a strain on the neuromuscular system. As a result, the wearable motion-assist device promotes the structural development and strengthening of neural loops, enabling treatment that activates the nervous system to maintain and improve the patient's motor function.

[0006] In order to understand the therapeutic effect and progress of a subject's motor dysfunction using a wearable motion assist device, the walking function of the subject is generally evaluated based on the walking distance and walking speed of the subject.

[0007] Regarding the evaluation of walking function, a gait measuring device has been proposed that detects the walking state of the person being measured, analyzes the walking state of the person being measured based on the walking data measured, and calculates the walking ability of the person being measured (see Patent Document 2). [Prior art documents] [Patent documents]

[0008] [Patent Document 1] Japanese Patent Application Laid-Open No. 2005-95561 [Patent Document 2] Japanese Patent Application Laid-Open No. 2015-130954 Summary of the Invention [Problem to be solved by the invention]

[0009] However, with conventional walking function evaluation methods, it is practically difficult to perform the evaluation after each treatment, given the burden on medical staff who must measure data while ensuring the safety of the subject, and it has been difficult to recognize changes over time in the subject's walking function following treatment using a wearable movement-assist device.

[0010] Furthermore, the method for evaluating walking function in Patent Document 2 also involves placing multiple sensors inside a walking mat, monitoring the position where the subject steps and the time from landing to lifting, and measuring the subject's stride length and walking speed, and simply evaluating walking function by analogy based on the timing of landing on the walking mat.

[0011] The present invention has been made in consideration of the above points, and provides a walking function evaluation device and a method for evaluating walking function that can recognize changes over time in a subject's walking function and significantly improve the speed with which a treatment plan for the subject can be created. The control program This is what we are trying to propose. [Means for solving the problem]

[0012] In order to solve the above problem, the present invention provides a wearable action-assist device that applies power to a subject according to each walking phase that constitutes the walking movement of the subject, covered A walking function evaluation device for evaluating the walking function of an examiner, outfit The wearable movement assist device is covered a drive unit that actively or passively drives in conjunction with the examiner's lower limb movement; covered a biosignal detection unit having a group of electrodes arranged on a body surface of the subject based on joints associated with lower limb movements of the subject, for detecting biopotential signals of the subject; raw Based on the biopotential signal acquired by the body signal detection unit, covered an optional control unit that generates power in the drive unit according to the examiner's will; Drive Based on the output signal from the actuator, covered a joint circumference detection unit that detects physical quantities around the joints associated with the examiner's lower limb movements; Seki Based on the physical quantity detected by the node surrounding detection unit, covered Identify each walking phase according to the examiner's walking task, each Power corresponding to the walking phase Drive an autonomous control unit that generates the Zui Intentional control unit and Self A control signal is synthesized from the control unit, and a driving current is generated according to the synthesized control signal. Drive a drive current generating unit that supplies the drive current to the driving unit; covered Based on the results of the floor reaction force sensor that detects the pressure distribution on the soles of the examiner's left and right feet, covered a gait synchronization calculation unit that calculates the gait period of the examiner; raw The biopotential signal detected by the body signal detection unit is Seki The physical quantity detected by the node surrounding detection unit and Pawn a signal normalization unit that normalizes the walking period calculated by the row synchronization calculation unit to a first signal pattern expressed in a plane coordinate system of time and amplitude for each walking period; Faith obtained from the code normalization part No.comparing the first signal pattern with a second signal pattern corresponding to a reference healthy individual; No. 1 signal pattern and No. a similarity calculation unit that quantitatively calculates the similarity between two signal patterns; kind Based on the similarity calculated by the similarity calculation unit, covered The present invention is provided with a walking function evaluation unit that evaluates the walking function of the examiner.

[0013] As a result, in a walking function evaluation device using a wearable motion assist device, by normalizing the first signal pattern of the subject's biopotential signal and the second signal pattern of the biopotential signal corresponding to a healthy subject and comparing the two, it becomes possible to recognize changes in the subject's walking function over time based on the similarity obtained from the comparison results.

[0014] Furthermore, in the present invention, a walking speed calculation unit is provided which calculates the stride length of the subject's walking movement based on the length of the subject's legs input in advance and the changes in physical quantities detected by the joint circumference detection unit, and calculates the walking speed of the subject based on the stride length and the walking period calculated by the walking synchronization calculation unit, and the walking function evaluation unit analyzes the correlation between the similarity calculated by the similarity calculation unit and the walking distance per specified time based on the walking speed calculated by the walking speed calculation unit.

[0015] As a result, the walking function evaluation device analyzes the correlation between the similarity between the first and second signal patterns and the walking distance per predetermined time based on the walking speed, and finds that the higher the similarity, the longer the walking distance per predetermined time. Therefore, it can be confirmed that the more similar the first signal pattern of a subject walking using a wearable motion-assist device is to the second signal pattern of a healthy subject, the longer the distance the subject can walk without using the device.

[0016] Furthermore, in the present invention, the similarity calculation unit uses differential dynamic time warping (DDTW) to compare the shapes of the first and second signal patterns over time and calculates the pattern similarity as similarity from the correspondence between the upward and downward trends. As a result, it can be confirmed that the smaller the DDTW value, the closer the first signal pattern of the subject is to the second signal pattern of the healthy subject.

[0017] Furthermore, in the present invention, a wearable action-assist device is used to apply power to a subject according to each walking phase constituting the walking movement of the subject, covered Gait function evaluation to evaluate the examiner's walking function Device control program In outfit The wearable movement assist device is covered a drive unit that is driven actively or passively in conjunction with the examiner's lower limb movements; covered Based on biopotential signals acquired from a body surface part of the subject based on a joint associated with the examiner's lower limb movement, covered Power according to the examiner's wishes Drive voluntary control exerted on the motor; Drive detected based on the output signal of the covered Based on the physical quantities around the joints caused by the examiner's lower limb movements, covered Identify each walking phase according to the examiner's walking task, each Power corresponding to the walking phase Drive The autonomous control generated by the moving part is synthesized, If In response to the control signal Drive Dynamic current Drive and supplying the fluid to the moving part, The control unit of the walking function evaluation device The body potential signal, Seki Physical quantities around the node and covered a first step of normalizing the gait cycle calculated based on the detection result of the pressure distribution on the soles of the examiner's left and right feet to a first signal pattern expressed in a plane coordinate system of time and amplitude for each gait cycle; No. Obtained from one step No. comparing the first signal pattern with a second signal pattern corresponding to a reference healthy individual; No. 1 signal pattern and No. a second step of quantitatively calculating the similarity between the two signal patterns; No. The third step is to evaluate the walking function of the subject based on the similarity calculated in the second step. Execute a series of processes consisting of I did so.

[0018] As a result, a walking function evaluation device using a wearable motion assist device Control program In this method, by normalizing the first signal pattern of the subject's biopotential signal and the second signal pattern of the biopotential signal corresponding to a healthy subject and comparing the two, it is possible to recognize the subject's walking function as a change over time based on the similarity obtained from the comparison results. [Effects of the Invention]

[0019] According to the present invention, a walking function evaluation device and a method for evaluating walking function of a subject are provided, which can significantly improve the speed of creating a treatment plan for the subject by recognizing changes in the walking function of the subject over time. The control program This can be achieved. [Brief explanation of the drawings]

[0020] [Figure 1] 1 is a conceptual diagram illustrating a walking assistance system according to an embodiment of the present invention. [Figure 2] 1 is a perspective view showing the external configuration of a wearable action-assist device according to an embodiment of the present invention; [Figure 3] 1 is a block diagram showing an internal configuration of a wearable action assist device according to an embodiment of the present invention. [Figure 4] FIG. 1 is a block diagram showing the internal configuration of a walking function evaluation device using a wearable movement assist device. [Figure 5] 1 is a chart showing information about subjects. [Figure 6] 10 is a graph showing a second signal pattern of a biopotential signal obtained from a right knee extensor muscle of a healthy subject. [Figure 7] 10 is a graph showing a second signal pattern of a biopotential signal obtained from a right knee extensor muscle of a healthy subject. [Figure 8]10 is a graph showing a normalized first signal pattern of a biopotential signal. [Figure 9] FIG. 10 is a schematic diagram illustrating an alignment state by DDTW. [Figure 10] 1 is a graph showing alignment results to which DDTW scores have been applied. [Figure 11] 10 is a graph showing a reference example of the results of DDTW. [Figure 12] 10 is a graph showing a reference example of the results of DDTW. [Figure 13] This is a chart combining 2MWT distance and DDTW score. [Figure 14] 1 is a diagram explaining an ANOVA table and correlation coefficients. DETAILED DESCRIPTION OF THE INVENTION

[0021] An embodiment of the present invention will be described in detail below with reference to the drawings.

[0022] (1) Configuration of the walking assistance system according to this embodiment 1 shows a walking assistance system 1 according to this embodiment. The walking assistance system 1 includes a wearable action-assist device 2 that assists the movement of a subject P, and a walking assistance device 3 that assists the subject P in rehabilitation through walking. The wearable action-assist device 2 and the walking assistance device 3 are connected to each other via wire or wirelessly so as to be able to communicate with each other.

[0023] First, the walking assistance device 3 is configured such that a pair of left and right frames 6L and 6R are curved and erected from the tip of the treadmill 5 on either side of the treadmill 5, and the subject P can grasp the end portions of both frames 6L and 6R with both hands.

[0024] The treadmill 5 has a walking belt 7 that moves in a circular motion due to the rotation of rollers. The rotation speed of the rollers can be changed in response to actuator drive, thereby changing the circulation speed of the walking belt 7.

[0025] The walking assistance device 3 has a subframe (not shown) that bridges between the left frame 6L and the right frame 6R that are erected from the treadmill 5, and is provided with a monitor 8, for example, a liquid crystal display, that displays the results of operations performed by the operating unit and various information necessary for assisting the subject in walking.

[0026] In this way, in the walking assistance system 1, the subject P wearing the wearable motion-assist device 2 can support rehabilitation through walking by holding one end of the pair of left frame 6L and right frame 6R of the walking assistance device 3 with both hands to stabilize his / her posture during walking.

[0027] (2) Configuration of the wearable action assist device according to this embodiment 2 shows a wearable action-assist device 2 according to this embodiment. The wearable action-assist device 2 is a device that applies power to a subject according to each walking phase that constitutes the walking motion of the subject, and operates by detecting bioelectrical signals (surface myoelectric potentials) that are generated when muscle force is generated by signals from the brain and the movement angles of the hip joints and knee joints of the wearer, and applying driving force from a drive mechanism based on these detection signals.

[0028] The lower limb type wearable action-assist device 2 in this embodiment includes a waist frame 10 attached to the waist of the subject, a lower limb frame 11 attached to the lower limbs of the wearer, a plurality of drive units 12L, 12R, 13L, 13R provided on the lower limb frame 11 corresponding to the joints of the wearer, cuffs 14L, 14R, 15L, 15R as assist force application members attached to the lower limb frame 11 so as to apply the forces of the drive units 12L, 12R, 13L, 13R to the wearer from the front or the back, a control device 30 (see FIG. 3 described later) that controls the drive units 12L, 12R, 13L, 13R based on signals resulting from the wearer's lower limb movements, a back unit 16 equipped with the control device, and an operation unit (not shown) used by a caregiver.

[0029] The control device 30 (FIG. 3) can drive the lower limb frames 11 relatively around the output axes of the actuators of the drive units 12L, 12R, 13L, and 13R corresponding to the joints of the subject. Each of the drive units 12L, 12R, 13L, and 13R is equipped with a group of sensors for detecting the drive torque and rotation angle of the actuator. The back unit 16 is equipped with a battery unit (not shown) for supplying power to drive the entire device.

[0030] The waist frame 10 is a member that is roughly C-shaped in plan view and opens forward to receive the waist of the subject and surround it from its rear to both left and right sides. It has a rear waist frame portion 17 located behind the subject, and a left waist frame portion 18L and a right waist frame portion 18R that extend forward while curving from both ends of the rear waist frame portion 17.

[0031] The left and right waist frame sections 18L, 18R are connected to the rear waist frame section 17 via an opening adjustment mechanism (not shown). The bases of the left and right waist frame sections 18L, 18R are inserted and held within the rear waist frame section 17 so as to be slidable in the left-right direction.

[0032] The lower limb frame 11 includes a right lower limb frame 19R attached to the right lower limb of the subject, and a left lower limb frame 19L attached to the left lower limb of the subject. The left lower limb frame 19L and the right lower limb frame 19R are formed symmetrically.

[0033] The left lower leg frame 19L includes a left thigh frame 20L positioned on the left side of the subject's left thigh, a left lower leg frame 21L positioned on the left side of the subject's left lower leg, and a left lower leg frame 22L on which the sole of the subject's left leg (or the sole of the left shoe, if shoes are worn) is placed. The left lower leg frame 19L is connected to the tip of the left waist frame 18L via a waist connecting mechanism 23L.

[0034] The right lower limb frame 19R includes a right thigh frame 20R located on the right side of the subject's right thigh, a right lower leg frame 21R located on the right side of the subject's right lower leg, and a right lower leg frame 22R on which the sole of the subject's right leg (or the sole of the right shoe, if wearing shoes) is placed. The right lower limb frame 21R is connected to the tip of the right waist frame 18R via a waist connection mechanism 23R.

[0035] The waist frame 10 (rear waist frame 17, right waist frame 18R, and left waist frame 18L) and the lower limb frame 11 (right lower limb frame 19R and left lower limb frame 19L) have a frame body formed in the shape of an elongated plate made of, for example, a metal such as stainless steel or carbon fiber, and are formed to be lightweight and highly rigid. In this embodiment, carbon fiber reinforced plastic (CFRP) and extra super duralumin, an aluminum alloy, are used as strength members.

[0036] The cuffs 14L, 14R, 15L, 15R are provided one each on the left thigh frame 20L, the right thigh frame 20R, the left lower leg frame 21L, and the right lower leg frame 21R.

[0037] The cuffs 14L, 14R (hereinafter referred to as "thigh cuffs") provided on the left thigh frame 20L and the right thigh frame 20R are supported by thigh cuff support mechanisms 24L, 24R attached to the lower ends of the thigh frame bodies. The thigh cuffs 14L, 14R have an arc-shaped curved attachment surface that can be fitted and placed against the subject's thigh. A fitting member is attached to the attachment surface of the thigh cuffs 14L, 14R so that they can fit tightly against the subject's thigh without any gaps.

[0038] The cuffs 15L, 15R (hereinafter referred to as "calf cuffs") provided on the left crus frame 21L and right crus frame 21R are supported by crus cuff support mechanisms 25L, 25R attached to the upper ends of the upper elements. The crus cuffs 15L, 15R have an arc-shaped attachment surface that can be fitted and placed against the subject's crus. A fitting member is attached to the attachment surface of the crus cuffs 15L, 15R so that they fit snugly against the subject's crus.

[0039] When the wearable action-assist device 2 is actually worn on a subject, dedicated shoes 26L, 26R are worn on the left and right feet, respectively, lower leg cuffs 15L, 15R are worn on the left and right lower legs, respectively, and thigh cuffs 14L, 14R are worn on the left and right thighs, respectively. Then, belts or the like are fastened to the shoes and cuffs so that the feet, lower legs, and thighs are integrated with the corresponding frames.

[0040] These dedicated shoes 26L, 26R consist of a pair of left and right shoes that hold the subject's feet in close contact from the toes to the ankles, and can measure load using a floor reaction force sensor (FRF sensor 60, described below) attached to the sole of the foot.

[0041] In this way, the wearable action-assist device 2 can control and assist walking movements based on bioelectric potential signals accompanying voluntary muscle activity according to the intention of the subject wearing the device.

[0042] (3) Internal system configuration of the wearable motion assist device Fig. 3 is a block diagram showing the configuration of the control system of the wearable action-assist device 2. As shown in Fig. 3, the control system 2X of the wearable action-assist device 2 includes a control device 30 that performs overall control of the entire system, a data storage unit 31 in which various data is stored in a database so as to be readable and writable in response to commands from the control device 30, and drive units 12L, 12R, 13L, and 13R that are actively or passively driven in conjunction with the movement of the lower limbs of the subject.

[0043] In addition, a potentiometer 32 that detects the rotation angle of the output shaft is provided coaxially with the output shaft of the actuator in the drive units 12L, 12R, 13L, and 13R, and is configured to detect the joint angle corresponding to the movement of the subject's lower limbs.

[0044] Furthermore, an absolute angle sensor 33 for measuring the absolute angle of the thigh relative to the vertical direction is mounted on the lower limb frame 11. This absolute angle sensor 33 is composed of an acceleration sensor and a gyro sensor, and is used for sensor fusion, which is a method of extracting new information using data from multiple sensors.

[0045] To calculate the absolute thigh angle, a first-order filter is used to remove the effects of translational motion and temperature drift in each sensor, and this first-order filter is calculated by adding weighted values ​​obtained from each sensor.

[0046] If the absolute angle of the thigh relative to the vertical direction is θabs(k), the angular velocity obtained by the gyro sensor is ω, the sampling period is dt, and the acceleration obtained by the acceleration sensor is α, then θabs(t) can be expressed as the following equation (1).

number

[0047] A biosignal detection unit 40 having a biosignal detection sensor (group of electrodes) is placed on the body surface of the subject (mainly the body surface of the thigh) based on the joint associated with the subject's lower limb movement, and is configured to detect bioelectric potential signals for moving the subject's knee joint.

[0048] The data storage unit 31 stores a command signal database 41 and a reference parameter database 42. The control device 30 is configured, for example, by a CPU (Central Processing Unit) chip having a memory, and includes an optional control unit 50, an autonomous control unit 51, a phase identification unit 52, and a gain change unit 53.

[0049] The optional control unit 50 controls the drive units 12L, 12R, 13L, and 13R to generate power according to the subject's will, based on the biopotential signal acquired by the biosignal detection unit 40. Specifically, the optional control unit 50 supplies a command signal corresponding to the detection signal of the biosignal detection unit 40 to the power amplification unit 54. The optional control unit 50 applies a predetermined command function f(t) or gain P to the biosignal detection unit 40 to generate the command signal. This gain P is a preset value or function, and can be adjusted via a gain change unit 53 using an external input.

[0050] It is also possible to select a method of controlling the drive torque (torque magnitude and rotation angle) of the actuator based on the angle data of the knee joint detected by the potentiometer 32. This method is effective when the degree of gait disorder associated with the subject's motor symptoms is relatively mild, or when the subject's skin is expected to be wet with sweat and there is a possibility that input of a biological signal from the biological signal detection unit 40 will not be obtained.

[0051] The data on the knee joint angle detected by the potentiometer 32, the data on the absolute angle of the thigh relative to the vertical direction detected by the absolute angle sensor 33, and the biosignal detected by the biosignal detection unit 40 are input into a reference parameter database 42.

[0052] Additionally, the soles of the pair of dedicated shoes 26L, 26R are provided with FRF (Floor Reaction Force) sensors 60 to detect the pressure distribution on the soles of the subject's left and right feet. The FRF sensors 60 can measure the load acting on the soles of the feet separately for the forefoot (toes) and the rearfoot (heel) separately.

[0053] This FRF sensor 60 may be composed of, for example, a piezoelectric element that outputs a voltage according to the applied load or a sensor whose capacitance changes according to the load, and can detect changes in load due to weight shift and whether or not the wearer's legs are in contact with the ground.

[0054] Furthermore, with the pair of dedicated shoes 26L, 26R, the center of gravity position can be determined from the balance of the loads on the soles of the left and right feet based on the detection results of each FRF sensor 60. In this way, with the pair of dedicated shoes 26L, 26R, it is possible to estimate to which side of the subject's left or right foot the center of gravity is biased, based on the data measured by each FRF sensor 60.

[0055] In addition to the shoe structure, each of the dedicated shoes 26L, 26R has an FRF sensor 60, an FRF control unit 61 consisting of an MCU (Micro Control Unit), and a transmitter 62. The output of the FRF sensor 60 is converted into a voltage via a converter 63, and then input to the FRF control unit 61 after high frequency bands are cut off via an LPF (Low Pass Filter) 64.

[0056] The FRF control unit 61 determines whether the subject has ground contact or not, as well as the load change caused by the weight shift of the subject, based on the detection results of the FRF sensor 60. The FRF control unit 61 wirelessly transmits the determined center of gravity position as FRF data via the transmission unit 62 to the receiving unit 65 in the device body.

[0057] The control device 30 receives the FRF data wirelessly transmitted from the transmitter 62 of each dedicated shoe 26L, 26R via the receiver 65, and then stores the load and center of gravity position on the soles of the left and right feet based on the FRF data in the reference parameter database 42 of the data storage unit 31.

[0058] The phase identification unit 52 compares the knee joint angle data detected by the potentiometer 32 and the load data detected by the FRF sensor 60 with the knee joint angle and load of the reference parameters stored in the reference parameter database 42. Based on the comparison result, the phase identification unit 52 identifies the phase of the subject's movement.

[0059] Then, when the autonomous control unit 51 obtains the control data for the phase identified by the phase identification unit 52, it generates a command signal according to the control data for this phase and supplies the command signal to the power amplification unit 54 to cause the drive units 12L, 12R, 13L, and 13R to generate this power.

[0060] Furthermore, the gain adjusted by the gain change unit 53 described above is input to the autonomous control unit 51, which generates a command signal according to this gain and outputs it to the power amplification unit 54. The power amplification unit 54 controls the current that drives the actuators of the drive units 12L, 12R, 13L, and 13R to control the magnitude of the torque and the rotation angle of the actuators, thereby applying an assist force from the actuators to the knee joint of the subject.

[0061] In this way, the autonomous control unit 51 identifies each walking phase according to the walking task of the subject based on the physical quantities detected by the joint circumference detection unit (potentiometer 32 and absolute angle sensor 33), and causes the drive units 12L, 12R, 13L, and 13R to generate power corresponding to each walking phase.

[0062] The power amplifier (drive current generator) 54 combines the control signals from the voluntary control unit 50 and the autonomous control unit 51, amplifies the drive current corresponding to the combined control signal, and supplies it to the actuators of the drive units 12L, 12R, 13L, and 13R. The torque of these actuators is transmitted as an assist force to the knee joint of the subject via the lower limb frame.

[0063] (4) Configuration of the walking function evaluation device according to this embodiment In the present invention, a walking function evaluation device 70 (FIG. 4, described later) using the wearable action assist device 2 described above is used to evaluate the walking function of a subject undergoing treatment.

[0064] As a premise, the biopotential signals of the lower limb muscles measured using the wearable motion-assist device 2 may be useful for evaluating the walking function of the subject, because the subject's muscle activity is measured every time walking therapy is performed using the wearable motion-assist device 2. The biopotential signals reflect changes in the subject's neuromuscular system caused by action potentials generated during movement control.

[0065] The signal patterns of bioelectrical signals obtained from the skin surface around the lower limb muscles change depending on muscle activity during walking.

[0066] During normal walking of a healthy person not wearing the wearable motion-assist device 2, the signal pattern of the biopotential signal will be characteristic of each measurement site. Similarly, the signal pattern of the biopotential signal of a subject wearing the wearable motion-assist device 2 is also thought to have characteristics, and analyzing this signal pattern will make it possible to record the activity of the neuromuscular system during walking. Furthermore, the relationship between the walking ability of the subject and the signal pattern of the biopotential signal measured when walking while wearing the wearable motion-assist device 2 may be applicable to the gait evaluation of subjects undergoing treatment.

[0067] For this reason, in the present invention, the signal pattern of the biopotential signal obtained from the subject undergoing treatment using the wearable action-assist device 2 is quantified, and evaluated by comparing it with the signal pattern of the biopotential signal corresponding to that of a healthy subject, thereby confirming the correlation between the signal pattern and the walking ability of the subject.

[0068] This walking function evaluation device 70 is a control system component provided in the control device 30 of the wearable action-assist device 2 described above, and as shown in FIG. 4, includes a walking synchronization calculation unit 71, a signal normalization unit 72, a similarity calculation unit 73, a walking function evaluation unit 74, and a walking speed calculation unit 75.

[0069] First, biopotential signals and data related to a walking test (gait cycle) are acquired from the subject using the wearable motion-assist device 2. Specifically, in treatment using the wearable motion-assist device 2 for a subject suffering from a progressive neuromuscular disease, the subject needs to walk for approximately 20 to 40 minutes per walking test. During this time, the wearable motion-assist device 2 measures biopotential signals obtained from the extensor and flexor muscles of the left and right knee joints and hip joints, and floor reaction forces (FRF data) of both legs as time-series data.

[0070] The results of time-series data measured during treatment using the wearable movement-assist device on seven subjects (Patient ID: A to G) with a progressive neuromuscular disease were used. The subjects also periodically performed a 2-minute walking test (2MWT), which measures the walking distance in 2 minutes, without wearing the wearable movement-assist device 2, in order to ensure accurate gait evaluation.

[0071] For seven subjects, walking tests were performed at a single facility within the past two years. The number of walking tests and 2MWT results performed during this study period varied for each subject. The disease type, gender, height, weight, and number of 2MWT results for each subject are shown in the table in Figure 5. In this table, MD, ALS, IBM, and SBMA represent muscular dystrophy, amyotrophic lateral sclerosis, inclusion body myositis, and spinal and femoral muscular atrophy, respectively.

[0072] The signal patterns of the biopotential signals obtained from these subjects are compared with the signal patterns of the biopotential signals obtained when a healthy person wearing the wearable action-assist device 2 is walking, to determine the similarity.

[0073] In the walking assistance system 1 shown in FIG. 1 described above, bioelectric potential signals obtained from the right knee extensor muscles of healthy subjects were measured while they walked on a treadmill 5 while wearing a wearable motion-assist device 2. Three healthy adult males aged 21 to 23 (Participants X to Z) were selected as the healthy subjects. The control parameters of the wearable motion-assist device 2 and the running speed of the walking belt 7 of the treadmill 5 were adjusted in advance to a walking speed comfortable for each participant.

[0074] The wearable motion-assist device 2 was used to measure and process the biopotential signals obtained from the right knee extensor muscles of participants X to Z. The average value of the signal patterns of the biopotential signals was then calculated for 90 walking cycles per participant, i.e., a total of 270 walking cycles.

[0075] The average value of this signal pattern was used as a standard for healthy subjects and was calculated using the following steps 1 to 5. The biopotential signal was divided into gait cycles starting from the moment of contact of the right heel, detected based on the values ​​of the floor reaction force sensor (FRF sensor 60) (Step 1). The biopotential signal was resampled to 101 points at regular intervals of 0 to 100 during each gait cycle and normalized by the duration of each gait cycle. The resampled values ​​were then interpolated using cubic spline interpolation (Step 2).

[0076] The amplitude of the biopotential signal was normalized for each gait cycle, with the maximum value set to 100 and the base value set to 0 (Step 3). For each sampling point, the average value of 270 biopotential signals was calculated and a signal pattern connecting these average values ​​was obtained (Step 4). The signal pattern obtained above was normalized together with the amplitude using the same method as in Step 3. This signal pattern was used as the gait standard for a healthy subject using the wearable motion-assist device 2 (Step 5).

[0077] Figures 6(A) and 6(B) and Figure 7(A) show the signal patterns of biopotential signals obtained from the right knee extensor muscles for healthy subjects (participants X to Z) walking on a treadmill 5 while wearing the wearable motion-assist device 2. Figure 7(B) shows the average value of a total of 270 walking patterns for the three participants, with the solid line and dashed line indicating the mean value and standard deviation range, respectively. The average pattern shown in Figure 7(B) was used as the reference biopotential signal pattern for healthy subjects to determine the similarity of the signal patterns between the subjects and healthy subjects.

[0078] In order to determine the above-mentioned walking period, the walking synchronization calculation unit 71 shown in FIG. 4 calculates the walking period of the subject based on the detection results of the floor reaction force sensor (FRF sensor 60) that detects the pressure distribution on the soles of the left and right feet of the subject.

[0079] The signal normalization unit 72 normalizes the biopotential signal detected by the biosignal detection unit 40 into a first signal pattern expressed in a plane coordinate system of time and amplitude for each walking cycle, based on the physical quantity detected by the joint circumference detection unit (potentiometer 32 and absolute angle sensor 33) and the walking cycle calculated by the walking synchronization calculation unit 71.

[0080] Specifically, in this embodiment, the wearable action-assist device 2 measures biopotential signals obtained from the right knee extensor muscle of a subject suffering from a progressive neuromuscular disease, and the signal pattern (first signal pattern) of the biopotential signals for each walking cycle is normalized with respect to amplitude and time and shown in FIGS. 8(A) and 8(B).

[0081] Figure 8(A) is a graph showing the normalized results of the measurement results from the first trial, and Figure 8(B) is a graph showing the normalized results of the measurement results three months later. These two normalized graphs are significantly different, and by comparing and analyzing the signal pattern of the biopotential signal obtained from a healthy subject (second signal pattern), it became possible to quantify the difference.

[0082] The similarity calculation unit 73 compares the first signal pattern obtained from the signal normalization unit 72 with a second signal pattern corresponding to a reference healthy subject, and quantitatively calculates the similarity between the first signal pattern and the second signal pattern. Specifically, the similarity calculation unit 73 uses differential dynamic time warping (DDTW) to compare the shapes of the first signal pattern and the second signal pattern in time series, and calculates the pattern similarity as similarity from the correspondence between the upward trend and the downward trend.

[0083] Therefore, we calculated similarity using a method (Derivative Dynamic Time Warping) proposed by EJ Keogh and MJ Pazzani of the University of California. Unlike specific comparison methods such as Pearson's correlation coefficient, root mean square error, and linear fit method, this method is a similarity calculation method that can relatively flexibly deal with time lags and nonlinear relationships between parameters.

[0084] The similarity between two time series data is calculated using the dynamic time warping (DDTW) algorithm as follows: First, let S be the first signal pattern obtained from the time series data of the subject, and let T be the second signal pattern obtained from the time series data of the healthy subject T. Let us assume the following equations (2) and (3):

number

number

[0085] In addition, to consider the shape of the pattern for evaluation, the first derivative of each time series was considered. For example, the differential estimate Ds[i] of a certain time series s can be expressed as the following equation (4). <i<Mとする。

number

[0086] All points of the first signal pattern S and the second signal pattern T were converted into the first signal pattern S' and the second signal pattern T' using the differential estimation formula expressed by formula (4), and then the arrays of the first signal pattern S' and the second signal pattern T' were aligned in consideration of the matrix, as shown in FIG. 9.

[0087] Each matrix element (i, j) belongs to the first signal pattern S' and the second signal pattern T' and corresponds to the alignment between points sj' and tj' having the values ​​of the differential estimates Ds[i] and Dt[j]. The correspondence between the first signal pattern S and the second signal pattern T and points sj and tj inherits the combination (i, j) that means the change in the differential value at points sj' and tj', and is expressed as the following equation (5).

number

[0088] Next, we constructed a warping path W, which is a continuous set of matrix elements that defines the correspondence between the first signal pattern S' and the second signal pattern T', to satisfy the following three conditions. First, the corner cells on the diagonal of the matrix are used as the starting and ending points. Second, the steps are limited to adjacent cells, including diagonally adjacent cells. Third, the points are arranged so that they do not monotonically decrease over time.

[0089] Since there are multiple warping paths W that satisfy the above three conditions, the optimal warping path W' is determined by minimizing the sum of d(si',tj') belonging to the matrix elements that include the warping path W, and then the DDTW value used to evaluate the similarity is calculated as shown in the following equation (6).

number

[0090] In this equation (6), L represents the length of W', and (s'ml, t'ml) represents the combination of l alignments of W'. The DDTW value represents the average difference in the derivatives of the two points S' and T' aligned by W'. Therefore, the smaller the DDTW value, the higher the similarity between the first signal pattern of the subject and the second signal pattern of the healthy subject.

[0091] As an example of expressing the similarity in the signal patterns of biopotential signals between a subject and a healthy subject when walking using the wearable action-assist device 2 using a DDTW score, the alignment results obtained by applying the DDTW score, which will be described later, to FIGS. 8(A) and 8(B) are shown in FIGS. 10(A) and 10(B).

[0092] The solid lines in Figures 10(A) and (B) show a first signal pattern of biopotential signals obtained from a subject undergoing treatment using the wearable action-assist device 2, which is similar to the signal pattern displayed in Figure 7(B). The dashed lines in Figures 10(A) and (B) show a second signal pattern of biopotential signals obtained from a healthy subject walking using the wearable action-assist device 2, which is similar to the signal pattern displayed in Figure 7(B). These results clearly show a significant difference between the first signal pattern of biopotential signals measured from the subject during the first treatment and the first signal pattern of biopotential signals measured three months later.

[0093] As can be seen from Figure 10(B), although the signal pattern of the biopotential signal is not shown in Figure 10(A), it reaches a maximum immediately after the initial contact with the floor, decreases towards the swing phase, and then increases during the swing phase. The muscle activity pattern shown in Figure 10(B) also resembles the gait of a healthy individual.

[0094] The lines connecting the graphs for the subject and healthy subjects in Figures 10(A) and (B) were aligned based on the DDTW score, and the similarity was determined using the differential change at each point between the lines. The DDTW value in Figure 10(A) was 2.56, and the DDTW value in Figure 10(B) was 0.96. Therefore, the DDTW value in Figure 10(B) is smaller, and it can be said that the first signal pattern of the subject observed in Figure 10(B) is closer to the second signal pattern of the healthy subject. Reference examples of DDTW results are shown in Figures 11(A) and (B) and Figures 12(A) and (B).

[0095] Next, the walking function evaluation unit 74 (FIG. 4) evaluates the walking function of the subject based on the similarity calculated by the similarity calculation unit 73. That is, a first signal pattern of biopotential signals obtained from the subject undergoing treatment using the wearable action-assist device 2 is compared with a second signal pattern of biopotential signals obtained from a healthy subject, and the correlation between the distance in the 2-minute walking test (2MWT) and the value of dynamic time warping (DDTW) is examined, thereby clarifying the relationship between the walking ability of the subject and the first signal pattern of biopotential signals.

[0096] Therefore, the walking speed calculation unit 75 calculates the stride length of the subject's walking movement based on the pre-inputted leg length of the subject and changes in physical quantities detected by the joint circumference detection unit (potentiometer 32 and absolute angle sensor 33), and calculates the walking speed of the subject based on the stride length and the walking period calculated by the walking synchronization calculation unit 71. Then, the walking function evaluation unit 74 analyzes the correlation between the similarity calculated by the similarity calculation unit 73 and the walking distance per predetermined time (2 minutes) based on the walking speed calculated by the walking speed calculation unit 75.

[0097] First, the "DDTW score" was calculated for the first signal pattern obtained from the biopotential signals during the treatment trial for seven subjects (patients A to G) using the wearable action-assist device 2. The DDTW score was expressed as the calculation result of the average of the DDTW values ​​expressed by the above-mentioned formula (5) for each test.

[0098] The 2MWT distance and DDTW score for each test were then combined to calculate a correlation coefficient. The 2MWT distance and DDTW score for each test for all subjects (patients A to G) are shown in the table in Figure 13. The correlation obtained from multiple measurements (number of tests) for each subject was coefficient can be interpreted as the correlation between subjects and the correlation within the subject itself. The correlation coefficient between subjects was evaluated as a weighted correlation coefficient, taking into account the various observations of each subject.

[0099] The procedure was carried out according to the calculation method by J.M.Bland and D.G. Altman (Calculating correlation coefficients with repeated observations). The actual weighted correlation coefficient WCC is expressed by the following formula (7):

number

[0100] Then, using multiple regression analysis (a statistical method for estimating the relationship between explanatory variables and dependent variables), the p-value, which is the cumulative probability of the occurrence of the t-value, was calculated from the F-test based on the t-value, which is the test statistic for a sample of seven subjects.Then, using multiple regression based on the calculation method by J.M.Bland and D.G. Altman mentioned above, the correlation coefficient within subjects was determined.

[0101] As shown in the table in Figure 13, the 2MWT distance was used as the outcome variable. The DDTW score and the subject, treated as a categorical factor using a dummy variable with six degrees of freedom, were used as predictor variables. An analysis of variance (ANOVA) table was used for regression, and the magnitude of the correlation coefficient (CCWP) within the subject can be expressed as the following equation (8).

number

[0102] The correlation between the 2MWT distance and the DDTW score was thus examined, and the inter-subject correlation coefficient was found to be -0.83. The t-value was 9.59, and the p-value was 2.08 × 10. Furthermore, multiple regression analysis was performed to determine the intra-subject correlation coefficient, resulting in the ANOVA table shown in Figure 14(A). Furthermore, the partial regression coefficient for the DDTW score had a negative sign. Therefore, the calculated intra-subject correlation coefficient was -0.39, and the corresponding p-value was 1.88 × 10.

[0103] As described above, the inter-subject data correlation was examined, and as shown in the table in Figure 14(B), the 2MWT distance and DDTW score showed a strong negative correlation coefficient of -0.83 (p = 2.08 × 10-4 < 0.01). Furthermore, the intra-subject data variability was examined, and a weak negative correlation coefficient of -0.39 (p = 1.88 × 10-2 < 0.05) was found between the 2MWT distance and DDTW score. These results confirmed that there is a significant relationship between the bioelectric signal pattern and walking ability while using a wearable mobility assist device.

[0104] In this way, the walking function evaluation device 70 using the wearable action assist device 2 normalizes the first signal pattern of the subject's biopotential signal and the second signal pattern of the biopotential signal corresponding to a healthy subject and compares the two, thereby making it possible to recognize changes in the subject's walking function over time based on the similarity obtained from the comparison results.

[0105] Furthermore, by analyzing the correlation between the similarity between the first and second signal patterns and the walking distance per predetermined time based on the walking speed, the walking function evaluation device 70 determines that the higher the similarity, the longer the walking distance per predetermined time. Therefore, it can be confirmed that the more similar the first signal pattern of a subject walking using the wearable action-assist device 2 is to the second signal pattern of a healthy subject, the longer the distance the subject can walk without using the device.

[0106] As a result, the walking function evaluation device 70 can recognize changes in the walking function of the subject over time, and can significantly improve the speed with which a treatment plan for the subject can be created.

[0107] (5) Other embodiments As described above, in this embodiment, the gait evaluation is performed mainly on the first signal pattern of the biopotential signal obtained from the subject's right knee extensor muscle, but the present invention is not limited to this, and a signal pattern of the biopotential signal obtained by integrating multiple muscles necessary for the subject's walking may also be applied to the gait evaluation.

[0108] In this embodiment, the subject is assisted in rehabilitation by walking on the treadmill 5 of the walking assistance device 3, but the present invention is not limited to this. The subject using the wearable motion-assist device 2 may also walk with a movable walker. [Explanation of symbols]

[0109] 1...walking assistance system, 2...wearable movement assistance device, 2X...control system, 3...walking assistance device, 5...treadmill, 6L...left frame, 6R...right frame, 7...walking belt, 8...monitor, 10...waist frame, 11...lower limb frame, 12L, 12R, 13L, 13R...drive unit, 26L, 26R...dedicated shoes, 30...control device, 31...data storage unit, 32...potentiometer, 33...absolute angle sensor, 40...biological signal detection unit, 41... Command signal database, 42...reference parameter database, 50...optional control unit, 51...autonomous control unit, 52...phase identification unit, 53...gain change unit, 54...power amplification unit, 60...FRF sensor, 61...FRF control unit, 62...transmitter, 63...converter, 64...LPF, 65...receiver, 70...walking function evaluation device, 71...walking synchronization calculation unit, 72...signal normalization unit, 73...similarity calculation unit, 74...walking function evaluation unit, 75...walking speed calculation unit.

Claims

1. 1. A walking function evaluation device for evaluating a walking function of a subject using a wearable motion assist device that applies power to the subject according to each walking phase constituting the walking movement of the subject, comprising: The wearable action-assist device includes: a drive unit that actively or passively drives the device in conjunction with the movement of the subject's lower limbs; a biosignal detection unit having a group of electrodes arranged on a body surface of the subject based on joints involved in lower limb movements of the subject, for detecting biopotential signals of the subject; an optional control unit that causes the drive unit to generate power according to the subject's will based on the biopotential signal acquired by the biosignal detection unit; a joint circumference detection unit that detects a physical quantity around the joint accompanying a lower limb movement of the subject based on an output signal from the drive unit; an autonomous control unit that identifies walking phases corresponding to the walking task of the subject based on the physical quantities detected by the joint circumference detection unit, and causes the drive unit to generate power corresponding to each walking phase; a drive current generating unit that combines control signals from the optional control unit and the autonomous control unit and supplies a drive current corresponding to the combined control signal to the drive unit; a gait synchronization calculation unit that calculates the gait cycle of the subject based on the detection results of a floor reaction force sensor that detects pressure distribution on the soles of the left and right feet of the subject; a signal normalization unit that normalizes the biopotential signal detected by the biosignal detection unit into a first signal pattern expressed in a plane coordinate system of time and amplitude for each gait cycle, based on the physical quantity detected by the joint circumference detection unit and the gait cycle calculated by the gait synchronization calculation unit; a similarity calculation unit that compares the first signal pattern obtained from the signal normalization unit with a second signal pattern corresponding to a reference healthy subject, and quantitatively calculates a similarity between the first signal pattern and the second signal pattern; a walking function evaluation unit that evaluates the walking function of the subject based on the similarity calculated by the similarity calculation unit; A walking function evaluation device comprising:

2. a walking speed calculation unit that calculates a stride length in a walking movement of the subject based on a leg length of the subject input in advance and a transition of a physical quantity detected by the joint circumference detection unit, and calculates a walking speed of the subject based on the stride length and a walking period calculated by the walking synchronization calculation unit, The walking function evaluation unit analyzes the correlation between the similarity calculated by the similarity calculation unit and the walking distance per predetermined time based on the walking speed calculated by the walking speed calculation unit. The walking function evaluation device according to claim 1 .

3. The similarity calculation unit uses differential dynamic time warping (DDTW) to compare shapes of the first signal pattern and the second signal pattern in time series and calculates the pattern similarity as the similarity from a correspondence between an upward trend and a downward trend.

3. The walking function evaluation device according to claim 1 or 2.

4. 1. A control program for a walking function evaluation device that evaluates a walking function of a subject using a wearable action-assist device that applies power to the subject according to each walking phase that constitutes the walking movement of the subject, comprising: the wearable action-assist device has a drive unit that is actively or passively driven in conjunction with a lower limb movement of the subject, and combines a voluntary control that causes the drive unit to generate a power according to the will of the subject based on a biopotential signal acquired from a body surface part of the subject with reference to a joint associated with the lower limb movement of the subject, and an autonomous control that identifies walking phases according to a walking task of the subject based on physical quantities around the joints associated with the lower limb movement of the subject detected based on an output signal from the drive unit, and causes the drive unit to generate a power corresponding to each walking phase, and supplies a drive current according to the combined control signal to the drive unit; A control unit of the walking function evaluation device a first step of normalizing the biopotential signals to a first signal pattern expressed in a plane coordinate system of time and amplitude for each gait cycle, the first signal pattern being based on a gait cycle calculated based on physical quantities around the joints and detection results of pressure distribution on the soles of the left and right feet of the subject; a second step of comparing the first signal pattern obtained in the first step with a second signal pattern corresponding to a reference healthy subject, and quantitatively calculating a similarity between the first signal pattern and the second signal pattern; a third step of evaluating the walking function of the subject based on the similarity calculated in the second step; A control program that executes a series of processes.

5. determining a stride length in a walking movement of the subject based on the leg length of the subject and the transition of the physical quantity around the joint, which are input in advance; and calculating a walking speed of the subject based on the stride length and the walking cycle; In the third step, a correlation between the similarity calculated in the second step and the walking distance per predetermined time based on the calculated walking speed is analyzed.

5. The control program according to claim 4.

6. In the third step, the shapes of the first signal pattern and the second signal pattern are compared in time series using differential dynamic time warping (DDTW), and pattern similarity is calculated as the similarity from a correspondence relationship between an upward trend and a downward trend.

6. The control program according to claim 4 or 5.

Images