Ankle joint rehabilitation training device

Abstract

An embodiment of the present application provides an ankle joint rehabilitation training device. The device obtains the actual motion parameters of a patient through a sensor assembly, and a controller determines a parameter difference based on the actual motion parameters and the applied motion parameters, and adjusts the applied motion parameters according to the parameter difference and the actual motion parameters. Finally, a motion instruction is generated, and a motor drives a footrest to move according to the motion instruction to perform rehabilitation training on the patient's foot. Since the current activity ability of the patient is considered during the rehabilitation training process, the rehabilitation training is more adaptable to the patient's own situation. At the same time, the rehabilitation training also considers the applied motion parameters corresponding to the target rehabilitation strategy. Therefore, while adapting to the patient himself, the rehabilitation efficiency is also taken into account, effectively improving the effect of the user's rehabilitation training.

Description

Technical Field

[0001] This application relates to the technical field of rehabilitation training, and particularly to an ankle joint rehabilitation training device. Background Art

[0002] Existing ankle joint rehabilitation devices usually adopt two methods. The first is that a rehabilitation therapist manually performs ankle joint rehabilitation training on a patient, and the second is to completely use an automated rehabilitation trainer to perform rehabilitation training on the patient's ankle joint.

[0003] For the device for manual rehabilitation treatment by a rehabilitation therapist, due to the low efficiency and high work intensity of manual rehabilitation by the rehabilitation therapist, and it is difficult for the rehabilitation therapist to accurately control the manipulation according to the patient's condition, the rehabilitation effects vary. When using an automated rehabilitation trainer to perform repetitive passive rehabilitation training on a patient, it is impossible to timely understand the patient's rehabilitation progress, lacking personalization and pertinence, which affects the rehabilitation effect. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide an ankle joint rehabilitation training device to improve the efficiency and rehabilitation effect of patient rehabilitation training.

[0005] In a first aspect, an embodiment of this application provides an ankle joint rehabilitation training device, which includes: a footrest, a bracket, a sensor assembly, a controller, and a motor. Among them, the footrest is arranged on the bracket, the footrest is movably connected to the bracket and can move axially relative to the bracket. The sensor assembly and the motor are arranged on the footrest, and both the sensor assembly and the motor are communicatively connected to the controller; the footrest is used to support the patient's foot; the sensor assembly is used to detect the actual motion parameters of the patient and send the actual motion parameters to the controller; among them, the actual motion parameters are used to characterize the patient's ankle joint motion ability; the controller is used to generate a motion instruction matching the patient according to the actual motion parameters and the application motion parameters corresponding to the patient's target rehabilitation strategy, and send the motion instruction to the motor; the controller further includes a comparison module and an adjustment module; among them, the comparison module is used to determine the parameter difference according to the actual motion parameters and the application motion parameters; the adjustment module is used to adjust the application motion parameters according to the actual motion parameters when the parameter difference is greater than the difference threshold; the motor is used to drive the footrest to move according to the motion instruction sent by the controller to perform rehabilitation training on the patient's foot.

[0006] Further, the above-mentioned actual motion parameters of the patient include the motion angle and current during the patient's movement; the sensor assembly includes an angle sensor and a motor current sensor; the angle sensor is used to obtain the motion angle of the patient during the movement; the motor current sensor is used to obtain the motor current corresponding to the motor during the patient's movement.

[0007] Further, the above-mentioned movement angles include at least one of the following: plantar flexion and dorsiflexion angles, adduction and abduction angles, and inversion and eversion angles; the angle sensors include at least one of the following: plantar flexion and dorsiflexion angle sensors, adduction and abduction angle sensors, and inversion and eversion angle sensors.

[0008] Further, the above-mentioned motor currents include at least one of the following: plantar flexion and dorsiflexion motor currents, adduction and abduction motor currents, and inversion and eversion motor currents; the motor current sensors include at least one of the following: plantar flexion and dorsiflexion motor current sensors, adduction and abduction motor current sensors, and inversion and eversion motor current sensors.

[0009] Further, the above-mentioned controller includes an angle conversion module and a current conversion module; wherein, the angle conversion module is used to perform coordinate transformation on the movement angle to obtain the joint range of motion of the patient; the current conversion module is used to obtain the muscle strength of the patient according to the measured current and the baseline current; the baseline current is the current generated during the movement of the ankle joint rehabilitation device when the patient does not exert force, and the measured current is the current generated during the movement of the ankle joint rehabilitation device when the patient exerts force.

[0010] Further, when adjusting the above-mentioned application movement parameters, let the optimization coefficient of each parameter be a i , i = 1, 2, 3...; wherein, x c is the value of the application movement parameter obtained by the first detection after the previous adjustment of the application movement parameter, and x d is the value of the application movement parameter obtained by the last detection after the previous adjustment of the application movement parameter; or wherein, x c is the value of the application movement parameter obtained by the first detection after the previous adjustment of the application movement parameter, and x d is the value of the application movement parameter obtained by the last detection after the previous adjustment of the application movement parameter, and x j is the secondary correction parameter corresponding to each application movement parameter obtained through machine learning, j = 1, 2, 3...; the application movement parameter is equal to the current actual movement parameter multiplied by a i .

[0011] Further, the above-mentioned controller further includes a machine learning module, which is used to input the actual movement parameter and the application movement parameter into a pre-trained machine learning model to obtain the corrected application movement parameter output by the machine learning model.

[0012] Further, the above-mentioned controller further includes a prescription selection module, which is used to obtain the application movement parameter from a preset prescription according to the actual movement parameter; wherein, the preset prescription includes at least one of the following: a preset movement action, a preset maximum movement angle, a preset number of action repetitions, and a preset action sequence.

[0013] Further, the above-mentioned controller further includes a teaching module, configured to obtain teaching data of the on-site teaching process of the rehabilitation therapist, and obtain application motion parameters matching the actual motion parameters according to the teaching data; wherein, the teaching data includes at least one of the angle, motion trajectory, angular velocity, and strength of the patient during the rehabilitation training of the patient by the rehabilitation therapist.

[0014] Further, the above-mentioned controller further includes an assisting module, configured to apply a preset force value in the same direction as the operation direction of the rehabilitation therapist to the motor during the on-site teaching process of the rehabilitation therapist, so as to drive the footrest to move by the motor; wherein, the net resistance received by the motor during the movement of the footrest is less than the preset force value.

[0015] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0016] The above-mentioned ankle joint rehabilitation training device provided by the embodiments of the present application obtains the actual motion parameters of the patient through the sensor assembly, determines the parameter difference according to the actual motion parameters and the application motion parameters by the controller, and adjusts the application motion parameters according to the parameter difference and the actual motion parameters, and finally generates a motion instruction, and the motor drives the footrest to move according to the motion instruction to perform rehabilitation training on the patient's foot. Since the current activity ability of the patient is considered during the rehabilitation training process, the rehabilitation training is more adaptable to the patient's own situation. At the same time, the rehabilitation training also considers the application motion parameters corresponding to the target rehabilitation strategy. Therefore, while adapting to the patient's own situation, it also takes into account the rehabilitation efficiency, effectively improving the effect of the user's rehabilitation training.

[0017] Other features and advantages of the present disclosure will be described in the following description, or some features and advantages can be inferred from the description or determined without doubt, or can be obtained by implementing the above technologies of the present disclosure.

[0018] In order to make the above-mentioned objects, features, and advantages of the present disclosure more obvious and understandable, the following specifically enumerates preferred embodiments and, in conjunction with the accompanying drawings, is described in detail as follows. Description of the Drawings

[0019] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the following will briefly introduce the drawings required for use in the description of the specific embodiments or the prior art. Obviously, the following drawings are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0020] Figure 1 It is a schematic diagram of the dorsiflexion and plantar flexion movements of the ankle joint;

[0021] Figure 2 Schematic diagram of adduction and abduction movements of the ankle joint;

[0022] Figure 3 Schematic diagram of inversion and eversion movements of the ankle joint;

[0023] Figure 4 Schematic structural diagram of an ankle joint rehabilitation training device provided by an embodiment of the present application;

[0024] Figure 5 Schematic structural diagram of an ankle joint rehabilitation training device in an actual application scenario provided by an embodiment of the present application;

[0025] Figure 6 Flowchart of an ankle joint rehabilitation training method provided by an embodiment of the present application. Detailed implementation manners

[0026] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions of the present application will be clearly and completely described below with reference to the accompanying drawings. Apparently, the described embodiments are some but not all of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without making creative efforts shall fall within the protection scope of the present application.

[0027] In the existing manual rehabilitation training method, the treatment efficiency of the therapist is low and the work intensity is high. At the same time, it is very difficult for the therapist to accurately control the manipulation according to the patient's condition, and the treatment and rehabilitation effects vary; when the therapist performs rehabilitation based on feel, it is not easy to achieve uniformity in strength, angle, and trajectory for multiple repeated movements in each direction. For different therapists treating the same patient, it is also difficult to ensure consistency in strength, angle, and trajectory. The existing automated rehabilitation trainers perform repeated passive rehabilitation training through manual parameter settings, and the rationality of parameter settings and training effects vary, and they cannot timely perceive the patient's rehabilitation progress, changes in range of motion, etc., and cannot adjust the training parameters adaptively according to the patient's rehabilitation progress. Based on this, the embodiments of the present application provide an ankle joint rehabilitation training device to improve the efficiency and rehabilitation effect of patient rehabilitation training.

[0028] First, the following will be combined with Figures 1-3 to explain several action types of the ankle joint involved in the embodiments of the present application.

[0029] Refer to Figure 1 , which shows the dorsiflexion and plantarflexion movements of the ankle joint. Dorsiflexion and plantarflexion are two terms used to describe the movement mode of the foot. Dorsiflexion refers to the movement in which the toes move upward and the dorsum of the foot approaches the front of the lower leg, that is, pulling up the toes. Plantarflexion refers to the movement in which the toes droop and the dorsum of the foot moves away from the front of the lower leg, that is, straightening the toes.

[0030] See Figure 2 , which shows the adduction and abduction movements of the ankle joint. Adduction and abduction are anatomical terms related to the possible movements of the joint. Adduction is the movement of the outer side of the foot towards the midline of the body, and abduction is the movement of the outer side of the foot away from the body. Both movements occur in a single plane of motion.

[0031] See Figure 3 , which shows the state of the inversion and eversion movements of the ankle joint.

[0032] The following provides a detailed introduction to an ankle joint rehabilitation training device provided by an embodiment of the present application. See Figure 4 The structural schematic diagram of an ankle joint rehabilitation training device according to an embodiment of the present application shown in, the device includes: a footrest 402, a bracket 404, a sensor assembly 406, a controller 408, and a motor 410. Among them, the footrest 402 is arranged on the bracket 404, the footrest 402 is movably connected to the bracket 404, and can move axially relative to the bracket 404. The sensor assembly 406 and the motor 410 are arranged on the footrest 402, and both the sensor assembly 406 and the motor 410 are communicatively connected to the controller 408.

[0033] Among them, the bracket 404 can be any support body that can be fixed on the ground or other planes. The footrest 402 is used to support the patient's foot. The footrest 402 can be designed in a form with adjustable size to adapt to the foot models of different patients, or can be designed in a fixed size, but can adapt to the foot models of most people. The specific structure of the footrest 402 is not limited in the embodiment of the present application.

[0034] The sensor assembly 406 is used to detect the actual motion parameters of the patient and send the actual motion parameters to the controller 408; among them, the actual motion parameters are used to characterize the ankle joint motion ability of the patient; each time rehabilitation training is carried out, the patient's motion ability will change. Correspondingly, the rehabilitation training for this patient also needs to be adjusted accordingly in order to better rehabilitate the patient. Therefore, the ankle joint rehabilitation training device provided by the embodiment of the present application detects the actual motion parameters of the patient through the sensor assembly 406 and sends them to the controller. In some examples, the actual motion parameters can include the angle of the patient's ankle joint, can also include the strength of the patient's ankle joint, or include the angular velocity of the patient's ankle joint during movement, etc. The motion parameters can be one of the above parameters, or can include all of the above parameters.

[0035] The controller 408 is configured to generate a motion command matching the patient based on the actual motion parameters and the applied motion parameters corresponding to the patient's target rehabilitation strategy, and send the motion command to the motor 410. Specifically, in addition to the above components and devices, the ankle rehabilitation training device provided by the embodiments of the present application may further include a memory, on which a plurality of rehabilitation strategies are stored. The plurality of rehabilitation strategies may be the teaching data of a rehabilitation therapist or the rehabilitation strategies selected by the patient during multiple rehabilitation training processes before the current moment. The patient's target rehabilitation strategy may be a rehabilitation strategy selected from a plurality of teaching data or historical rehabilitation strategies, or a rehabilitation strategy automatically selected by the controller according to the patient's previous motion condition.

[0036] In some possible implementation manners, the above controller includes an angle conversion module and a current conversion module; wherein, the angle conversion module is configured to perform coordinate system transformation on the motion angle to obtain the joint range of motion of the patient; the current conversion module is configured to obtain the muscle strength of the patient according to the measured current and the baseline current; the baseline current is the current generated during the motion of the ankle rehabilitation training device when the patient does not exert force, and the measured current is the current generated during the motion of the ankle rehabilitation training device when the patient exerts force.

[0037] After the plantar flexion / dorsiflexion angle, adduction / abduction angle, inversion / eversion angle, plantar flexion / dorsiflexion current, adduction / abduction current, and inversion / eversion current detected by the sensor assembly, the controller analyzes and calculates the above angles and currents.

[0038] For the angles, the plantar flexion / dorsiflexion angle, adduction / abduction angle, and inversion / eversion angle are transformed into the joint range of motion of the patient's ankle joint through coordinate system transformation.

[0039] For the motor current, the plantar flexion / dorsiflexion current, adduction / abduction current, and inversion / eversion current are converted into the patient's muscle strength through calculation. Specifically, the patient's muscle strength can be calculated using the following formula:

[0040] Patient muscle strength N = (measured current Ix - baseline current I0) torque coefficient K1 * muscle strength coefficient K2.

[0041] The motor 410 is configured to drive the footrest 402 to move according to the motion command sent by the controller 408, so as to perform rehabilitation training on the patient's ankle joint. After receiving the motion command from the controller 408, the motor 410 can control the footrest 402 to perform corresponding motions. Specifically, the motor 410 may be composed of a plantar flexion / dorsiflexion motor, an adduction / abduction motor, an inversion / eversion motor, and a motor controller. After receiving the command from the controller 408, the motor controller analyzes and calculates and then converts it into parameters for controlling the motor. The parameters for controlling the motor include the motion speed, angle range, output torque magnitude, etc. in each motion direction. The motor controller controls the motor to perform motion control according to the motion trajectory, angle change, and force magnitude change of the data parameters.

[0042] The above-mentioned ankle joint rehabilitation training device provided by the embodiments of the present application obtains the actual motion parameters of the patient through the sensor assembly, determines the parameter difference according to the actual motion parameters and the applied motion parameters by the controller, adjusts the applied motion parameters according to the parameter difference and the actual motion parameters, finally generates a motion instruction, and drives the footrest to move by the motor according to the motion instruction to perform rehabilitation training on the patient's foot. Since the current activity ability of the patient is considered during the rehabilitation training process, the rehabilitation training is more adaptable to the patient's own situation. At the same time, the rehabilitation training also considers the applied motion parameters corresponding to the target rehabilitation strategy. Therefore, while adapting to the patient's own situation, it also takes into account the rehabilitation efficiency, effectively improving the effect of the user's rehabilitation training.

[0043] In some possible implementation manners, the actual motion parameters of the patient include the motion angle and current of the patient during the motion; in order to detect the motion angle and current of the patient, the above-mentioned sensor assembly may specifically include an angle sensor and a motor current sensor;

[0044] Among them, the angle sensor is used to obtain the motion angle of the patient during the motion; and the motion angle includes at least one of the following: plantar flexion and dorsiflexion angle, adduction and abduction angle, and inversion and eversion angle; based on this, the above-mentioned angle sensor includes at least one of the following: plantar flexion and dorsiflexion angle sensor, adduction and abduction angle sensor, and inversion and eversion angle sensor.

[0045] The working principle of angle measurement is that the patient places the leg on the footrest, and the controller notifies the motor to drive the patient to perform periodic directional motions of the plantar flexion and dorsiflexion angle, adduction and abduction angle, and inversion and eversion angle. The executed periodic directional motions can be set according to experience. For example, the period can be set to 5.

[0046] During the motion, the motion angles of the patient are collected in real time through the plantar flexion and dorsiflexion angle sensor, adduction and abduction angle sensor, and inversion and eversion angle sensor. The controller collects the motor current through the plantar flexion and dorsiflexion motor current sensor, adduction and abduction motor current sensor, and inversion and eversion motor current sensor. When the current increases, it is judged that the patient has moved to his own limit angle, and the system records the motion angle at this time. For example, the threshold value for the increase in current can be set to: more than 50% beyond the normal operation.

[0047] The motor current sensor is used to obtain the motor current corresponding to the motor during the motion of the patient. The motor current includes at least one of the following: plantar flexion and dorsiflexion motor current, adduction and abduction motor current, and inversion and eversion motor current; based on this, the above-mentioned motor current sensor includes at least one of the following: plantar flexion and dorsiflexion motor current sensor, adduction and abduction motor current sensor, and inversion and eversion motor current sensor.

[0048] The working principle of current measurement is that the controller notifies the motor to work in the position mode. After the patient places the leg on the footrest, the plantar flexion / dorsiflexion motor current sensor, adduction / abduction motor current sensor, and inversion / eversion motor current sensor are used to collect and record the current parameters of plantar flexion / dorsiflexion, adduction / abduction, and inversion / eversion at this time, which are the baseline currents.

[0049] As Figure 5 shown, it is a schematic structural diagram of a specific application of an ankle joint rehabilitation training device provided by an embodiment of the present application. The device includes a sole motor 501, an ankle side motor 502, a footrest 503, a calf support 504, and a bracket 505. The device also includes a sensor assembly (not shown in the figure) and a controller (not shown in the figure). As Figure 5 shown, Figure 5 The ankle joint rehabilitation training device in

[0050] is the position when the user is in a lying position. During the actual rehabilitation training process, each motor drives the patient to actively exert force in the directions of plantar flexion / dorsiflexion, adduction / abduction, and inversion / eversion respectively. During this process, the plantar flexion / dorsiflexion, adduction / abduction, and inversion / eversion motor current sensors are used to collect and record the current parameters of plantar flexion / dorsiflexion, adduction / abduction, and inversion / eversion, which can be called the measured current.

[0051] From the above description, it can be seen that the baseline current is the current generated when the patient does not actively exert force, while the measured current is the current generated when the patient actively exerts force. By comparing the measured current and the baseline current, the magnitude of the current generated by the person's own strength can be obtained, which corresponds to the magnitude of the person's exertion.

[0052] Before the ankle joint rehabilitator starts the rehabilitation training, the step of selecting the data file corresponding to the target rehabilitation strategy can be carried out first. For example, the teaching data of the current rehabilitation therapist, that is, the teaching rehabilitation training data file, can be selected. It is also possible to select the teaching data file generated by other rehabilitation therapists on this device or remote similar devices, that is, the prescription data.

[0053] Regardless of whether the teaching data or the prescription data is selected, the controller will perform data file decryption and data calibration identification. Only when the identification result is a safe and valid data file can it be used. When the data file identification is abnormal, it will be prohibited from being used.

[0054] For the case of the user selecting a prescription in the above two situations, the above-mentioned controller in the embodiment of the present application further includes a prescription selection module, which is used to obtain application motion parameters from a preset prescription according to actual motion parameters; wherein, the preset prescription includes at least one of the following: a preset motion action, a preset maximum motion angle, a preset number of action repetitions, and a preset action sequence.

[0055] In the prescription mode, for example, when the patient performs the dorsiflexion and plantar flexion movements 10 times repeatedly, the amplitude, or rather the maximum range of motion, usually gradually increases. For example, if the patient's maximum range of motion is 20 degrees, then the prescription may be to gradually increase from 15 degrees to slightly more than 20 degrees, such as 22 degrees. If the patient's range of motion increases to 25 degrees, the corresponding range of motion of the prescription may become the range of 18 - 28 degrees. It will slightly exceed the patient's limit.

[0056] For the case of the user selecting teaching data in the above two situations, the above-mentioned controller in the embodiment of the present application further includes a teaching module, which is used to obtain the teaching data of the on-site teaching process of the rehabilitation therapist and obtain the application motion parameters matching the actual motion parameters according to the teaching data; wherein, the teaching data includes at least one of the patient's angle, motion trajectory, angular velocity, and strength during the rehabilitation training process of the rehabilitation therapist for the patient.

[0057] The teaching mode refers to the teaching of the ankle joint rehabilitation training device by the rehabilitation therapist, that is: the controller of the ankle joint rehabilitation training device records the actions of the rehabilitation therapist driving the foot support, and then drives the patient's ankle to move according to the recorded actions. The rehabilitation therapist uses their clinical experience and techniques to hold the foot support with their hands and perform any one or a combination of degrees of freedom movements of plantar flexion and dorsiflexion, adduction and abduction, inversion and eversion, in order to operate easily and autonomously and achieve a real experience similar to the rehabilitation therapist operating the patient's ankle joint.

[0058] Since the ankle joint rehabilitation training device provided in the embodiment of the present application has motors at each joint, when the rehabilitation therapist performs teaching, it is necessary to overcome the resistance of the motors to drive the device to move, which is relatively laborious. Therefore, it is necessary to configure each motor to actively conform to the actions of the rehabilitation therapist after sensing the thrust, so as to achieve the effect of saving the physical strength of the rehabilitation therapist. Based on this, the above-mentioned controller further includes an assistance module, which is used to apply a preset force value in the same direction as the operation direction of the rehabilitation therapist to the motor during the on-site teaching process of the rehabilitation therapist, so that the motor drives the foot support to move; wherein, the net resistance received by the motor during the movement of the foot support is less than the preset force value.

[0059] For example, when the controller detects that the rehabilitation therapist is performing teaching work, the operating force of the rehabilitation therapist can operate the parameter movement within the range of >= 5N. That is, the resistance felt by the rehabilitation therapist during the hand movement will not exceed 5N. That is, it moves in the direction of the hand movement of the rehabilitation therapist, thereby assisting in eliminating the resistance of the device itself. To achieve the effect of saving the power consumption of the rehabilitation therapist.

[0060] In the rehabilitation training mode, when the patient is undergoing rehabilitation training, the controller will collect and record the angular changes of plantar flexion and dorsiflexion, adduction and abduction, and inversion and eversion. At the same time, the controller periodically collects the current changes of plantar flexion and dorsiflexion, adduction and abduction, and inversion and eversion through the plantar flexion and dorsiflexion motor current sensor, adduction and abduction motor current sensor, and inversion and eversion motor current sensor, and saves the records. For example, data can be collected every 10 ms.

[0061] After obtaining the above-mentioned angle and current data, in order to apply the data more accurately to the patient's rehabilitation training process, the data can also be filtered, that is, filter the data parameters of the plantar flexion and dorsiflexion angle, adduction and abduction angle, inversion and eversion angle, plantar flexion and dorsiflexion current, adduction and abduction current, and inversion and eversion current collected and recorded, and remove the outliers generated by current oscillation, the outliers generated by mechanical structure jitter, and the artificial jitter outliers of the rehabilitation therapist's operation.

[0062] After selecting the data file, the controller will automatically adjust the parameters of the data file according to the measurement data feedback by the measurement data system of the patient, and adjust them to the joint range of motion angle range and muscle strength range parameters matching the patient's measurement.

[0063] At the same time, the controller provides the function for the rehabilitation therapist to manually modify the parameters. The rehabilitation therapist can set the time period for rehabilitation training using the data file, for example, set the cycle range from 1 to 100 times. After selecting and setting the parameters, transmit them to the controller.

[0064] After the patient has undergone several rehabilitation trainings, the strength and range of motion of the foot will recover to some extent. Since the sensor component measures the actual motion parameters of the patient each time, at this time, the actual motion parameters will be different from the motion parameters previously set by the ankle robot. As the patient continues to recover, this difference will become larger and larger. Therefore, when the difference is greater than a certain value, it is necessary to adjust the applied motion parameters of the patient to ensure the effect of rehabilitation training.

[0065] Based on this, in some possible implementation manners, the above-mentioned controller further includes a machine learning module, which is used to input the actual motion parameters and the applied motion parameters into a pre-trained machine learning model to obtain the corrected applied motion parameters output by the machine learning model.

[0066] Specifically, when the patient is undergoing rehabilitation training, the machine learning module will learn the actual motion parameters of the patient in real time. The actual motion parameters include the motion trajectory, the magnitude of the patient's active force, and the motion speed.

[0067] When the difference shows significant statistical significance, such as being greater than a set difference threshold, the machine learning module will perform computational analysis to propose optimized rehabilitation training parameters, and reset the applied motion parameters according to the optimized rehabilitation training parameters, thus forming an intelligent and automated intelligent rehabilitation training loop system.

[0068] When adjusting the applied motion parameters, let the optimization coefficient of each parameter be a i , where i = 1, 2, 3...; where x c is the value of the applied motion parameter obtained from the first detection after the previous adjustment of the applied motion parameter, and x d is the value of the applied motion parameter obtained from the last detection after the previous adjustment of the applied motion parameter.

[0069] Preferably, where x c is the value of the applied motion parameter obtained from the first detection after the previous adjustment of the applied motion parameter, x d is the value of the applied motion parameter obtained from the last detection after the previous adjustment of the applied motion parameter, x j is the secondary correction parameter corresponding to each applied motion parameter obtained through machine learning, where j = 1, 2, 3... x j is the further correction of a i . Since a i only considers the change in ankle joint movement ability during the previous correction period, there are certain limitations.

[0070] The secondary correction parameter is obtained through machine learning. The training set of the machine learning model is the motion parameter values and applied parameter values of different patients in previous times. There are two training objectives. One is to make the increase in the actual motion parameter during the next training period as large as possible after a single adjustment of the applied parameter. The other is to make the increase in the final actual motion parameter as large as possible after all adjustments of the applied parameters. After being verified by the test set, the secondary correction coefficient of a i is obtained when modifying the applied parameter value each time. For example, from the start of training, when making the first modification of the applied parameter, a 1 , and its corresponding secondary correction coefficient is x 1 , and so on. Since the secondary correction parameter considers the historical effects and cumulative effects of previous modifications, as well as the non-linear characteristics of the rehabilitation curve, the effect of each modification of the applied parameter value is optimized. The applied motion parameter is equal to the current actual motion parameter multiplied by (1 + a i ).

[0071] In the above embodiments, by continuously optimizing the application motion parameters of the patient using a machine learning model during the training process, the application motion parameters are always matched with the current physical condition of the patient, saving the rehabilitation time and effectively improving the effect of the rehabilitation training.

[0072] In some possible implementation manners, the above controller may further include a teaching data module, and the teaching data module specifically includes: a function curve generator and a storage module.

[0073] The therapist operates the movement and can arbitrarily change the force magnitude, movement trajectory, and speed. During the whole process, the perception acquisition system periodically acquires the angular changes of plantar flexion / dorsiflexion, adduction / abduction, and inversion / eversion during the whole operation process of the rehabilitation therapist through the plantar flexion / dorsiflexion angle sensor, adduction / abduction angle sensor, and inversion / eversion angle sensor, and saves the records (acquired angles, movement trajectories, angular velocities, and forces. As the basis for the subsequent control of the ankle joint robot to reproduce the actions of the rehabilitation therapist).

[0074] During the whole process, the controller periodically acquires the current changes of plantar flexion / dorsiflexion, adduction / abduction, and inversion / eversion through the plantar flexion / dorsiflexion motor current sensor, adduction / abduction motor current sensor, and inversion / eversion motor current sensor, and saves the records. For example, it can be acquired once every 10 ms.

[0075] Therefore, the function curve generator can generate a function curve through the following steps:

[0076] A11. Calculate the movement speed based on the acquisition time period of 10 ms and the movement angle change data.

[0077] A12. Generate a combined movement trajectory curve of plantar flexion / dorsiflexion, adduction / abduction, and inversion / eversion according to the effective technical parameters of the plantar flexion / dorsiflexion angle, adduction / abduction angle, inversion / eversion angle, plantar flexion / dorsiflexion current, adduction / abduction current, and inversion / eversion current after filtering.

[0078] A13. Apply the current changes of the effective plantar flexion / dorsiflexion current, adduction / abduction current, and inversion / eversion current to the movement trajectory curve. Since when the ankle joint rehabilitation training device drives the human foot to move, in addition to the changes in angle and angular velocity, it is also necessary to change the magnitude of the force applied to the foot. Therefore, in addition to the angle parameters, parameters of current changes are also required to control the force exerted by the motor. The current change is the change in the manual force of the rehabilitation therapist under teaching.

[0079] A14. Finally, generate a function curve with movement trajectory, speed change, and force change.

[0080] The data storage module stores data specifically through the following process:

[0081] A21. Extract data from the function curve generated by the function curve generator, that is, extract a set of data at regular intervals. These extracted data will form a set of rehabilitation movements of the ankle joint rehabilitation training device provided in the embodiments of the present application, which can be stored for subsequent reuse or shared with other ankle joint rehabilitation training devices for use.

[0082] A22. After encrypting and calibrating the data, convert it into a data file stored in a computer.

[0083] The generated data file supports local single-machine use and also supports being copied or downloaded remotely via the network to intelligent rehabilitation training devices elsewhere, realizing multi-machine sharing of the data file.

[0084] Based on the above ankle joint rehabilitation training device, the embodiments of the present application further provide an ankle joint rehabilitation training method, which is applied to a controller. The controller is the controller in the above ankle joint rehabilitation training device. Refer to Figure 6 the flowchart of the ankle joint rehabilitation training method shown, and this method specifically includes the following steps:

[0085] S602: Detect the actual movement parameters of the patient through the sensor component; wherein, the actual movement parameters are used to characterize the movement ability of the patient's ankle joint at the current moment;

[0086] Among them, the actual movement parameters of the patient include the movement angle and current during the patient's movement; in order to detect the movement angle and current of the patient, the above sensor component may specifically include an angle sensor and a motor current sensor;

[0087] Among them, the angle sensor is used to obtain the movement angle of the patient during the movement; and the movement angle at least includes the following three: plantar flexion and dorsiflexion angle, adduction and abduction angle, and inversion and eversion angle; based on this, the above angle sensor at least includes the following three: plantar flexion and dorsiflexion angle sensor, adduction and abduction angle sensor, and inversion and eversion angle sensor.

[0088] The motor current sensor is used to obtain the motor current corresponding to the motor during the patient's movement. The motor current at least includes the following three: plantar flexion and dorsiflexion motor current, adduction and abduction motor current, and inversion and eversion motor current; based on this, the above motor current sensor at least includes the following three: plantar flexion and dorsiflexion motor current sensor, adduction and abduction motor current sensor, and inversion and eversion motor current sensor.

[0089] S604: Generate a movement instruction matching the patient according to the actual movement parameters and the application movement parameters corresponding to the target rehabilitation strategy selected by the patient;

[0090] Specifically, after obtaining the actual motion parameters, the controller can compare them with the application motion parameters corresponding to the target rehabilitation strategy selected by the patient, and generate a motion instruction according to the comparison result.

[0091] First, the controller performs coordinate transformation on the motion angle to obtain the joint range of motion of the patient.

[0092] Furthermore, the motor current can include the measured current and the baseline current. The muscle strength of the patient can be obtained based on the measured current and the baseline current. Among them, the baseline current is the current generated during the movement of the ankle joint rehabilitation training device when the patient does not exert force, and the measured current is the current generated during the movement of the ankle joint rehabilitation training device when the patient exerts force.

[0093] The muscle strength of the patient can be calculated by the following formula:

[0094] Patient muscle strength N = (Measured current Ix - Baseline current I0) Torque coefficient K1 * Muscle strength coefficient K2. The target rehabilitation strategy can be the teaching data of the current rehabilitation therapist selected by the user, that is, the teaching rehabilitation training data file. It can also be the teaching data file selected by the user from other rehabilitation therapists generated on this device or remote similar devices, that is, the prescription data.

[0095] S606: Trigger the motor to execute the motion instruction, so that the motor drives the footrest to move, in order to perform rehabilitation training on the patient's foot.

[0096] In the above-mentioned ankle joint rehabilitation training method provided by the embodiments of the present application, the actual motion parameters of the patient are obtained through the sensor component, the motion instruction is generated by the controller according to the actual motion parameters and the application motion parameters, and finally the motor drives the footrest to move according to the motion instruction to perform rehabilitation training on the patient's foot. Since the current activity ability of the patient is considered during the rehabilitation training process, the rehabilitation training is more adaptable to the patient's own situation. At the same time, the rehabilitation training also considers the application motion parameters corresponding to the target rehabilitation strategy. Therefore, while adapting to the patient's own situation, it also takes into account the rehabilitation efficiency, effectively improving the effect of the user's rehabilitation training.

[0097] In order to better perform rehabilitation training on the patient, step S604 in the above method, generating a motion instruction matching the patient according to the actual motion parameters and the application motion parameters corresponding to the target rehabilitation strategy selected by the patient, can specifically include:

[0098] (1) Determine the parameter difference according to the actual motion parameters and the application motion parameters;

[0099] (2) When the parameter difference is greater than the difference threshold, adjust the application motion parameters according to the actual motion parameters.

[0100] Specifically, when the patient is undergoing rehabilitation training, the machine learning module will learn the actual motion parameters of the patient in real time. The actual motion parameters include the motion trajectory, the magnitude of the patient's active force, and the motion speed.

[0101] When the difference shows significant statistical significance, for example, greater than the set difference threshold, the machine learning module will perform calculation and analysis to propose optimized rehabilitation training parameters, and reset the applied motion parameters according to the optimized rehabilitation training parameters. In this way, an intelligent and automated intelligent rehabilitation training loop system is formed.

[0102] In the description of the present application, it should be noted that the orientation or positional relationship indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation to the present application. In addition, the terms "first", "second", "third" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance.

[0103] Finally, it should be noted that the above-described embodiments are only specific embodiments of the present application, which are used to illustrate the technical solutions of the present application, rather than to limit it. The protection scope of the present application is not limited thereto. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that any person skilled in the art within the technical scope disclosed by the present application can still modify the technical solutions recorded in the foregoing embodiments, or can easily think of changes, or make equivalent replacements for some of the technical features; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. An ankle joint rehabilitation training device, characterized in that, the device comprises: a footrest, a bracket, a sensor assembly, a controller and a motor. Among them, the footrest is arranged on the bracket, the footrest is movably connected to the bracket and can move axially relative to the bracket. The sensor assembly and the motor are arranged on the footrest, and both the sensor assembly and the motor are communicatively connected to the controller; the sensor assembly includes an angle sensor and a motor current sensor; the angle sensor is used to obtain the movement angle of the patient during the movement process; the motor current sensor is used to obtain the motor current corresponding to the motor during the movement process of the patient; the footrest is used to support the patient's foot; the sensor assembly is used to detect the actual movement parameters of the patient and send the actual movement parameters to the controller; among them, the actual movement parameters are used to characterize the ankle joint movement ability of the patient; the actual movement parameters of the patient include the movement angle and current of the patient during the movement process; the controller is used to generate a movement instruction matching the patient according to the actual movement parameters and the application movement parameters corresponding to the patient's target rehabilitation strategy, and send the movement instruction to the motor; the controller further includes a machine learning module, a prescription selection module, a teaching module, an assistance module, a comparison module and an adjustment module; among them, the comparison module is used to determine the parameter difference according to the actual movement parameters and the application movement parameters; The adjustment module is used to adjust the application motion parameters according to the actual motion parameters when the parameter difference is greater than the difference threshold; wherein, when adjusting the application motion parameters, the optimization coefficient of each parameter is set as a i , i = 1, 2, 3...; wherein, x c is the value of the application motion parameter detected for the first time after the previous adjustment of the application motion parameters, and x d is the value of the application motion parameter detected for the last time after the previous adjustment of the application motion parameters; or wherein, x c is the value of the application motion parameter detected for the first time after the previous adjustment of the application motion parameters, and x d is the value of the application motion parameter detected for the last time after the previous adjustment of the application motion parameters, and x j is the secondary correction parameter corresponding to each application motion parameter obtained through machine learning, j = 1, 2, 3...; the application motion parameter is equal to the current actual motion parameter multiplied by a i ; the motor is used to drive the footrest to move according to the movement instruction sent by the controller, so as to perform rehabilitation training on the patient's foot.

2. The device according to claim 1, characterized in that, the movement angle includes at least one of the following: plantar flexion and dorsiflexion angle, adduction and abduction angle, and inversion and eversion angle; the angle sensor includes at least one of the following: plantar flexion and dorsiflexion angle sensor, adduction and abduction angle sensor, and inversion and eversion angle sensor.

3. The device according to claim 1, characterized in that, the motor current includes at least one of the following: plantar flexion and dorsiflexion motor current, adduction and abduction motor current, and inversion and eversion motor current; the motor current sensor includes at least one of the following: plantar flexion and dorsiflexion motor current sensor, adduction and abduction motor current sensor, and inversion and eversion motor current sensor.

4. The device according to claim 1, characterized in that, the controller includes an angle conversion module and a current conversion module; among them, the angle conversion module is used to perform coordinate system transformation on the movement angle to obtain the joint range of motion of the patient; the current conversion module is used to obtain the muscle strength of the patient according to the measured current and the baseline current; the baseline current is the current generated during the movement process of the ankle joint rehabilitation device when the patient does not exert force, and the measured current is the current generated during the movement process of the ankle joint rehabilitation device when the patient exerts force.

5. The device according to claim 1, characterized in that, The machine learning module is configured to input the actual motion parameters and the applied motion parameters into a pre-trained machine learning model, and obtain the corrected applied motion parameters output by the machine learning model.

6. The device according to claim 1, wherein, the prescription selection module is configured to obtain the applied motion parameters from a preset prescription according to the actual motion parameters; wherein, the preset prescription includes at least one of the following: a preset motion action, a preset maximum motion angle, a preset number of action repetitions, and a preset action sequence.

7. The device according to claim 1, wherein, the teaching module is configured to obtain teaching data of the on-site teaching process of the rehabilitation therapist, and obtain the applied motion parameters matching the actual motion parameters according to the teaching data; wherein, the teaching data includes at least one of the angle, motion trajectory, angular velocity, and strength of the patient during the rehabilitation training process of the rehabilitation therapist for the patient.

8. The device according to claim 7, wherein, the assistance module is configured to apply a preset force value in the same direction as the operation direction of the rehabilitation therapist to the motor during the on-site teaching process of the rehabilitation therapist, so that the motor drives the footrest to move; wherein, the net resistance received by the motor during the movement of the footrest is less than the preset force value.

Images