Spasticity rehabilitation auxiliary training method and device after brain injury, terminal and medium

Abstract

This invention provides a method, device, terminal, and medium for assistive training in the rehabilitation of spasticity after brain injury. The method includes: acquiring surface electromyography (EMG) signals and the start time of postural transition; obtaining average amplitude, maximum voluntary contraction amplitude, left-right symmetry index, stability index, and pre-activation time; obtaining a core muscle group function score based on the calculated average amplitude, maximum voluntary contraction amplitude, left-right symmetry index, stability index, and pre-activation time; determining the patient's training position, training goals, training mode, initial treatment parameters, and treatment optimization goals; guiding the patient in adaptive rehabilitation training, collecting surface EMG signals of the patient's core muscle groups in real time, dynamically adjusting stimulation parameters through an adaptive prediction model and the surface EMG signals, and completing the assistive rehabilitation training for spasticity after brain injury after the adaptive prediction model converges. This reduces reliance on the therapist's subjective experience and achieves precise rehabilitation training.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, specifically to a method, device, terminal, and medium for assistive training in the rehabilitation of spasticity after brain injury. Background Technology

[0002] Spasticity following brain injury is a common sequela, characterized by increased muscle tone, hyperreflexia, and abnormal movement patterns, severely impacting patients' quality of life. Current rehabilitation methods largely focus on passive traction, functional electrical stimulation, or repetitive motion training of the extremities. While these methods have some effect, they often neglect the "proximal-priority" control principle of human movement. The core muscle groups, acting as the "engine" and stabilizer of human movement, directly influence motor control and spasticity inhibition in distal limbs. Research shows that by activating and stabilizing the core muscle groups, the neural control mechanism of "proximal stability ensuring distal flexibility" can effectively inhibit abnormal spasticity patterns in the limbs.

[0003] However, current rehabilitation training for spasticity after brain injury mainly relies on the therapist's subjective experience, making it difficult to achieve precise rehabilitation training. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention proposes a method, device, terminal, and medium for assistive training in the rehabilitation of spasticity after brain injury, aiming to reduce reliance on the therapist's subjective experience and achieve precise rehabilitation training.

[0005] In a first aspect, embodiments of this application provide an auxiliary training method for the rehabilitation of spasticity after brain injury, the method comprising: Acquire surface electromyography signals and the start time of position transitions under multiple body positions; Based on the start time of the body position change and the surface electromyography signal, the average amplitude, maximum voluntary contraction amplitude, left-right symmetry index, stability index, and pre-activation time are obtained. The core muscle group function score is obtained based on the calculated average amplitude, the maximum voluntary contraction amplitude, the left-right symmetry index, the stability index, and the pre-activation time. Based on core muscle group function scores, left-right symmetry indicators, stability indicators, and pre-activation time, determine the patient's training position, training goals, training mode, initial treatment parameters, and treatment optimization goals. Based on the patient's training position, training goals, training mode, initial treatment parameters, and optimization goals, the patient is guided to undergo adaptive rehabilitation training. Surface electromyography (EMG) signals of the patient's core muscle groups are collected in real time. Stimulation parameters are dynamically adjusted using an adaptive prediction model and the EMG signals. After the adaptive prediction model converges, the patient completes auxiliary rehabilitation training for spasticity following brain injury.

[0006] Optionally, the step of obtaining the average amplitude, maximum voluntary contraction amplitude, left-right symmetry index, stability index, and pre-activation time based on the body position change start time and the surface electromyography signal includes: The surface electromyography signals within the stable body position segment were rectified using a full-wave rectifier to obtain the root mean square (RMS) data. The average amplitude is obtained based on the root mean square value data; For each body position, a maximum voluntary isometric contraction test was performed, and the maximum amplitude value in the root mean square value data was selected to obtain the maximum voluntary contraction amplitude. The amplitudes of the left and right muscle groups were filtered from the root mean square data, and the left-right symmetry index was obtained based on the amplitudes of the left and right muscle groups. Based on the root mean square value data and the average amplitude, a stability index is obtained; The pre-activation time is obtained based on the start time of body position change and the onset time of electromyographic activation in the surface electromyographic signal.

[0007] Optionally, the step of performing full-wave rectification on the electromyographic signal within the stable postural phase to obtain root mean square (RMS) data includes: ; in, This is the root mean square value at time t; The value of the i-th surface electromyography signal; is the window length for surface electromyography signals.

[0008] Optionally, obtaining the average amplitude based on the root mean square value data includes: ; in, Let J be the root mean square values ​​of the j windows; This represents the number of sliding windows within the stable segment.

[0009] Optionally, the step of filtering the amplitudes of the left and right muscle groups from the root mean square data, and obtaining the left-right symmetry index based on the amplitudes of the left and right muscle groups, includes: ; in, It is an indicator of left-right symmetry; The mean RMS value of the left-side muscle of the same name in the stable segment; The mean RMS value of the corresponding muscle on the right side within the stable segment.

[0010] Optionally, the step of determining the variance of the amplitude signal within the stable segment from the root mean square (RMS) data to obtain a stability index includes: ; in, For stability indicators; This represents the total number of root mean square values ​​within the stable segment. Let be the root mean square value of the j-th window; The average amplitude.

[0011] Optionally, obtaining the pre-activation time based on the body position change start time and the electromyographic activation start time in the surface electromyographic signal includes: ; in, This is the pre-activation time; This refers to the onset time of electromyographic activation. This is the start time of the body position change.

[0012] Optionally, the step of obtaining a core muscle group function score based on the calculated average amplitude, the maximum voluntary contraction amplitude, the left-right symmetry index, the stability index, and the pre-activation time includes: ; in, Assess core muscle function. Weighting based on body position; Position the patient correctly; for Average amplitude under body position; for Maximum voluntary contraction amplitude under body position; The first evaluation coefficient; It is an indicator of left-right symmetry; This is the second evaluation coefficient; For stability indicators; The maximum acceptable variance; This is the third evaluation coefficient; This is the pre-activation time; To activate the reference time.

[0013] Optionally, based on core muscle function scores, bilateral symmetry indicators, stability indicators, and pre-activation time, the patient's training position, training goals, training mode, and initial treatment parameters are determined, including: Positions with low core muscle function scores were selected to determine the patient's training positions. Based on left-right symmetry indicators, stability indicators, and pre-activation time, the patient's training goals are obtained; Select the stimulation device corresponding to the muscle group that corresponds to the patient's training goal to obtain the patient's training mode; Initial treatment parameters and treatment optimization objectives are determined using a multi-objective optimization function.

[0014] Optionally, the real-time acquisition of surface electromyography (EMG) signals of the patient's core muscle groups, the dynamic adjustment of stimulation parameters through an adaptive prediction model and the EMG signals, and the completion of post-brain injury spasticity rehabilitation auxiliary training after the adaptive prediction model converges, include: In each control cycle k, the current electromyographic signal is acquired, and the real-time electromyographic amplitude is extracted; Predict future electromyographic responses using the current adaptive prediction model; The optimization problem is solved by rolling optimization of the objective function to determine the treatment parameters for the next time step; By updating the model parameters using recursive least squares with a forgetting factor, the model gradually adapts to the patient's individual response characteristics, and completes the auxiliary training for post-brain injury spasticity rehabilitation after the adaptive prediction model converges.

[0015] Secondly, embodiments of this application provide a rehabilitation assistive training device, comprising: The data acquisition module is used to acquire surface electromyography signals and the start time of position transitions in multiple body positions. The index determination module is used to obtain the average amplitude, maximum voluntary contraction amplitude, left-right symmetry index, stability index, and pre-activation time based on the body position change start time and the surface electromyography signal. The scoring module is used to obtain a core muscle group function score based on the calculated average amplitude, the maximum voluntary contraction amplitude, the left-right symmetry index, the stability index, and the pre-activation time. The training parameter determination module is used to determine the patient's training position, training goals, training mode, initial treatment parameters, and treatment optimization goals based on core muscle group function scores, left-right symmetry indicators, stability indicators, and pre-activation time. The training adjustment and determination module is used to guide the patient to conduct adaptive rehabilitation training based on the patient's training position, the patient's training goal, the patient's training mode, the initial treatment parameters, and the treatment optimization goal. It collects surface electromyography (EMG) signals of the patient's core muscle groups in real time, dynamically adjusts stimulation parameters through an adaptive prediction model and the surface EMG signals, and completes auxiliary rehabilitation training for spasticity after brain injury after the adaptive prediction model converges.

[0016] Thirdly, embodiments of this application provide a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the brain injury spasticity rehabilitation auxiliary training method as described in any one of the first aspects above.

[0017] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the brain injury spasticity rehabilitation auxiliary training method as described in any one of the first aspects above.

[0018] Fifthly, embodiments of this application provide a computer program product that, when run on a terminal device, causes the terminal device to execute the brain injury spasticity rehabilitation auxiliary training method described in any of the first aspects above.

[0019] In this embodiment, surface electromyography (EMG) signals and postural transition start times are acquired under multiple body position states. Based on the postural transition start times and the EMG signals, average amplitude, maximum voluntary contraction amplitude, left-right symmetry index, stability index, and pre-activation time are obtained. A core muscle group function score is obtained based on the calculated average amplitude, maximum voluntary contraction amplitude, left-right symmetry index, stability index, and pre-activation time. Based on the core muscle group function score, left-right symmetry index, stability index, and pre-activation time, the patient's training position, training goals, training mode, initial treatment parameters, and treatment optimization goals are determined. The patient is guided to undergo adaptive rehabilitation training based on the patient's training position, training goals, training mode, initial treatment parameters, and treatment optimization goals. Real-time acquisition of EMG signals from the patient's core muscle groups is used. Stimulation parameters are dynamically adjusted using an adaptive prediction model and the EMG signals. After the adaptive prediction model converges, post-brain injury spasticity rehabilitation auxiliary training is completed. This reduces reliance on the therapist's subjective experience and achieves precise rehabilitation training. Attached Figure Description

[0020] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0021] Figure 1 This is a flowchart illustrating the first embodiment of the auxiliary training method for spasticity rehabilitation after brain injury provided in this application. Figure 2 This is a schematic diagram of the structure of the rehabilitation assistive training device provided in the embodiments of this application; Figure 3 This is a schematic diagram of the structure of the terminal device provided in the embodiments of this application. Detailed Implementation

[0022] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0023] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0024] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0025] Figure 1 The illustration shows a flowchart of a first embodiment of the rehabilitation auxiliary training method for spasticity after brain injury provided in this application. This is an example and not a limitation; the method can be applied to the aforementioned rehabilitation auxiliary training device. Figure 1 As shown, the method may include: S10, acquire surface electromyography signals and the start time of body position transition under multiple body position states; To reduce reliance on the therapist's subjective experience and achieve precise rehabilitation training, the rehabilitation assistive training device acquires surface electromyographic signals and the start time of positional transitions in multiple body positions.

[0026] Before acquiring surface electromyography signals and the start time of position transitions under multiple body positions, the process may include: Send posture adjustment instructions via language; Furthermore, based on the body position adjustment command, surface electromyography (EMG) signals and the start time of body position transitions were obtained under multiple body position states.

[0027] S20, based on the start time of the body position change and the surface electromyography signal, the average amplitude, maximum voluntary contraction amplitude, left-right symmetry index, stability index and pre-activation time are obtained; After acquiring surface electromyography (EMG) signals and the start time of body position transition, the rehabilitation assistive training device obtains the average amplitude, maximum voluntary contraction amplitude, left-right symmetry index, stability index, and pre-activation time based on the start time of body position transition and the surface EMG signals.

[0028] As one implementation method, based on the start time of the body position change and the surface electromyography signal, the average amplitude, maximum voluntary contraction amplitude, left-right symmetry index, stability index, and pre-activation time are obtained, specifically including: A1, full-wave rectification of the surface electromyography signal within the stable body position segment to obtain root mean square value data; After obtaining the surface electromyography (EMG) signal, the rehabilitation assistive training device performs full-wave rectification on the surface EMG signal within the stable body position segment to obtain the root mean square (RMS) value data. As one implementation method, the surface electromyography signal within the stable postural phase is subjected to full-wave rectification to obtain root mean square (RMS) data, including: ; in, This is the root mean square value at time t; The value of the i-th surface electromyography signal; denoted as the window length for the surface electromyography signal, where the window length N corresponds to a time of 100 ms (i.e., 200 sampling points) and the sliding step size is 50 ms.

[0029] A2, based on the root mean square value data, the average amplitude is obtained; After obtaining the root mean square (RMS) data, the rehabilitation assistive training device calculates the average amplitude based on the RMS data.

[0030] As one implementation method, the average amplitude is obtained based on the root mean square (RMS) data, including: ; in, Let J be the root mean square values ​​of the j windows; This represents the number of sliding windows within the stable segment.

[0031] A3. Perform maximum voluntary isometric contraction tests on each body position, and select the maximum amplitude from the root mean square value data to obtain the maximum voluntary contraction amplitude. After obtaining the root mean square (RMS) data, the rehabilitation assistive training device performs a maximum voluntary isometric contraction test for each body position. The device then selects the maximum amplitude from the RMS data to obtain the maximum voluntary contraction amplitude. Before formal evaluation, a maximum voluntary isometric contraction test is performed for each body position. For example, in a supine position, the patient attempts to extend their abdomen against resistance; in a seated position, they attempt to extend their trunk backward. The maximum RMS value during the test (the peak value from three tests) is recorded as the maximum voluntary contraction amplitude for that body position. .

[0032] A4. Filter the amplitude of the left and right muscle groups from the root mean square value data, and obtain the left-right symmetry index based on the amplitude of the left and right muscle groups. After obtaining the root mean square (RMS) data, the rehabilitation assistive training device filters the amplitude of the left and right muscle groups from the RMS data, and obtains the left-right symmetry index based on the amplitude of the left and right muscle groups.

[0033] In other words, after obtaining the root mean square (RMS) data, the rehabilitation assistive training device calculates the mean amplitude of the left and right muscle groups within the stable segment, and obtains the left-right symmetry index based on the amplitude of the left and right muscle groups.

[0034] As one implementation method, the amplitudes of the left and right muscle groups are screened from the root mean square (RMS) data, and based on the amplitudes of the left and right muscle groups, a left-right symmetry index is obtained, including: ; in, It is an indicator of left-right symmetry; The mean RMS value of the left-side muscle of the same name in the stable segment; The mean RMS value of the corresponding muscle on the right side within the stable segment.

[0035] A5. Based on the root mean square value data and the average amplitude, the stability index is obtained; After obtaining the root mean square (RMS) value data and the average amplitude, the rehabilitation assistive training device obtains a stability index based on the RMS value data and the average amplitude.

[0036] As one implementation method, a stability index is obtained based on the root mean square value data and the average amplitude, including: ; in, For stability indicators; This is the total number of root mean square values ​​within the stable segment, which is also the number of windows; Let be the root mean square value of the j-th window; The average amplitude is given. A smaller stability index (variance) indicates more stable muscle activation; a larger stability index (variance) suggests tremors or instability.

[0037] A6. The pre-activation time is obtained based on the start time of body position change and the onset time of electromyographic activation in the surface electromyographic signal.

[0038] After obtaining the start time of body position change and the surface electromyography (EMG) signal, the rehabilitation assistive training device obtains the pre-activation time based on the start time of body position change and the onset time of EMG activation in the surface EMG signal.

[0039] As one implementation method, the pre-activation time is obtained based on the start time of body position change and the onset time of electromyographic activation in the surface electromyographic signal, specifically including: ; in, This is the pre-activation time; This refers to the onset time of electromyographic activation. To enhance the detection sensitivity of activation moments by calculating TKEO on the electromyographic signals of the transition segment, a threshold was then set, and activation was determined to begin when the TKEO sequence continuously exceeded the threshold. This refers to the start time of the body position change. This is to detect the start time of the body position change using an IMU.

[0040] like If the value is less than or equal to zero, the muscle has been pre-activated before the change of body position; if... If the value is greater than zero, the muscle response is delayed. S30, based on the calculated average amplitude, the maximum voluntary contraction amplitude, the left-right symmetry index, the stability index, and the pre-activation time, the core muscle group function score is obtained; The rehabilitation assistive training device obtains the average amplitude, maximum voluntary contraction amplitude, left-right symmetry index, stability index, and pre-activation time. Based on the calculated average amplitude, maximum voluntary contraction amplitude, left-right symmetry index, stability index, and pre-activation time, it obtains the core muscle group function score.

[0041] As one implementation method, a core muscle group function score is obtained based on the calculated average amplitude, the maximum voluntary contraction amplitude, the left-right symmetry index, the stability index, and the pre-activation time, specifically including: ; in, Assess core muscle function. Weighting based on body position; Position the patient correctly; for Average amplitude under body position; for Maximum voluntary contraction amplitude under body position; The first evaluation coefficient; It is an indicator of left-right symmetry; This is the second evaluation coefficient; For stability indicators; The maximum acceptable variance; This is the third evaluation coefficient; This is the pre-activation time; To activate the reference time.

[0042] S40, based on core muscle group function scores, left-right symmetry indicators, stability indicators, and pre-activation time, determine the patient's training position, training goals, training mode, and initial treatment parameters. After obtaining the core muscle group function score, left-right symmetry index, stability index, and pre-activation time, the rehabilitation assistive training device determines the patient's training position, training goals, training mode, and initial treatment parameters based on the core muscle group function score, left-right symmetry index, stability index, and pre-activation time.

[0043] As one implementation method, based on core muscle group function scores, left-right symmetry indicators, stability indicators, and pre-activation time, the patient's training position, training goals, training mode, and initial treatment parameters are determined, specifically including: B1. Select the patient's training position by screening for the position with the lowest core muscle function score; After obtaining the core muscle function score, the rehabilitation assistive training device selects the body positions with lower core muscle function scores to obtain the patient's training positions.

[0044] In other words, after obtaining the core muscle function score, the rehabilitation assistive training device prioritizes training in positions with lower core muscle function scores. For example, if the core muscle function score is low in the standing position, the focus is on training core control in the standing position. Similarly, if the core muscle function score is low in the lateral decubitus position, the focus is on training core control in the standing position.

[0045] B2, based on left-right symmetry indicators, stability indicators, and pre-activation time, the patient's training goals are obtained; After obtaining the left-right symmetry index, stability index, and pre-activation time, the rehabilitation assistive training device selects the index with the largest difference in standard deviation between the corresponding target values ​​of the left-right symmetry index, stability index, and pre-activation time to obtain the training index. The training index is then used to train the patient to obtain the training goal.

[0046] In other words, after obtaining left-right symmetry indicators, stability indicators, and pre-activation time, the rehabilitation assistive training device sets training goals targeting the patient's weaknesses. This involves selecting parameters that are relatively related to the left-right symmetry indicators. Parameters relative to stability indices And parameters related to pre-activation time The indicator with the largest difference in standard deviation between the target value and the target value is used to obtain the training indicator. The training indicator is then used to train the patient to obtain the training target.

[0047] In other words, after obtaining the left-right symmetry index, stability index, and pre-activation time, the rehabilitation assistive training device, if the parameters of the left-right symmetry index are relatively... If the value is too low, the goal is to increase the left-right electromyography ratio to above 0.9; if the stability index is relatively low... If the value is too low, the goal is to reduce the electromyographic variance to the normal range; if the pre-activation time is a relatively low parameter... If it is too low, the goal is to shorten the pre-activation time.

[0048] B3. Select the stimulation device corresponding to the muscle group that corresponds to the patient's training goal to obtain the patient's training mode. After obtaining the core muscle group function score, the rehabilitation assistive training device selects the stimulation device corresponding to the muscle group corresponding to the patient's training goal to obtain the patient's training mode.

[0049] In other words, single or combined stimulation modes and corresponding stimulation devices are selected based on the patient's condition. Stimulation methods can include: providing deep proprioceptive input through abdominal and chest strap inflation to enhance core muscle perception; applying rhythmic vibrations to activate muscle spindles and induce muscle contraction; providing real-time electromyographic signals to the patient in visual or auditory form to guide active activation of core muscles; and combining virtual reality or games to guide the patient to complete goal-oriented movements in specific postures.

[0050] B4 determines the initial treatment parameters and treatment optimization objectives through a multi-objective optimization function.

[0051] After obtaining the core muscle group function score, the rehabilitation assistive training device determines the initial treatment parameters and treatment optimization goals through a multi-objective optimization function.

[0052] In other words, after obtaining the core muscle group function score, the rehabilitation assistive training device uses a multi-objective optimization function to determine the initial stimulation parameters (such as inflation pressure, vibration frequency, feedback threshold, etc.). Treatment optimization goals can include efficacy goals, safety goals, and comfort goals. The efficacy goal is to achieve the predicted electromyographic response. Approaching the target value The safety objective is to minimize stimulus intensity. Comfort objectives include Wie parameters, which reflect patient preferences during holidays. .

[0053] That is, the initial treatment parameters are obtained by predicting the electromyographic response under given parameters through a pre-trained machine learning model (such as Gaussian process regression) and solving the optimization problem.

[0054] S50, based on the patient's training position, the patient's training goal, the patient's training mode, the initial treatment parameters, and the treatment optimization goal, guide the patient to conduct adaptive rehabilitation training, collect the surface electromyography signals of the patient's core muscle groups in real time, dynamically adjust the stimulation parameters through the adaptive prediction model and the surface electromyography signals, and complete the auxiliary training for spasticity rehabilitation after brain injury after the adaptive prediction model converges. After obtaining the patient's training position, training goals, training mode, initial treatment parameters, and treatment optimization goals, the rehabilitation assistive training device guides the patient to conduct adaptive rehabilitation training based on the patient's training position, training goals, training mode, initial treatment parameters, and treatment optimization goals. It collects surface electromyography (EMG) signals of the patient's core muscle groups in real time, dynamically adjusts stimulation parameters through an adaptive prediction model and the surface EMG signals, and completes the rehabilitation assistive training for spasticity after brain injury after the adaptive prediction model converges.

[0055] As one implementation method, surface electromyography (EMG) signals of the patient's core muscle groups are acquired in real time. Stimulation parameters are dynamically adjusted using an adaptive prediction model and the surface EMG signals. After the adaptive prediction model converges, auxiliary rehabilitation training for spasticity following brain injury is completed, specifically including: C1, in each control cycle k, acquires the current electromyographic signal and extracts the real-time electromyographic amplitude. ; After acquiring the patient's training position, training goals, training mode, initial treatment parameters, and treatment optimization goals, the rehabilitation assistive training device collects the current electromyographic signal and extracts the real-time electromyographic amplitude in each control cycle k. .

[0056] C2, using the current adaptive prediction model to predict the electromyographic response at future moments; After acquiring the patient's training position, training goals, training mode, initial treatment parameters, and treatment optimization goals, the rehabilitation assistive training device uses the current adaptive prediction model to predict the electromyographic response at future moments.

[0057] As one implementation method, the electromyographic response at future moments is predicted using a current adaptive prediction model, specifically including: ; in, The predicted electromyographic response is the expected amplitude (or processed electromyographic characteristics, such as RMS value) of the surface electromyographic signal of the core muscle group at the next control cycle k+1. This value is the therapeutic indicator that the system hopes to achieve. For control cycle number; For example, the treatment parameter vector at the current moment; This is a regression vector containing historical input and output data, used to capture the dynamic characteristics of the system; A predictive model is a machine learning or system identification model used to predict future electromyographic responses based on current parameters and historical data; it can be a linear regression model.

[0058] C3 solves the optimization problem by using a rolling optimization objective function to determine the treatment parameters for the next time step. ; After predicting the electromyographic response at future moments using the current adaptive prediction model, the rehabilitation assistive training device solves the optimization problem by rolling optimization objective function to determine the treatment parameters for the next moment. ; As one implementation method, the optimization problem is solved by rolling optimization of the objective function to determine the treatment parameters for the next time step, specifically including: in, Let be the objective function, representing the cost that needs to be minimized to balance efficacy, safety, and comfort; To predict the time domain, i.e., to predict how many future moments in advance, the parameter sequence for the next N steps will be optimized simultaneously. The predicted electromyographic response at time k+j is given by the prediction model. It is obtained by recursion; The target electromyographic response at time k+j is the desired ideal electromyographic level. It is a norm; Weighting coefficients; These are treatment parameters for future moments, and the decision variables to be optimized. The parameter vector that provides the patient with the greatest comfort.

[0059] C4 updates the model parameters using recursive least squares with a forgetting factor, allowing the model to gradually adapt to the patient's individual response characteristics, and completes the auxiliary training for post-brain injury spasticity rehabilitation after the adaptive prediction model converges.

[0060] After solving the optimization problem by rolling the objective function and determining the treatment parameters for the next moment, the rehabilitation assistive training device updates the model parameters by using a recursive least squares method with a forgetting factor, so that the model gradually adapts to the patient's personalized response characteristics, and completes the rehabilitation assistive training for spasticity after brain injury after the adaptive prediction model converges.

[0061] Monitor electromyographic signals in real time to see if they exceed the safe threshold. If excessive contraction or discomfort occurs, immediately reduce the stimulation intensity or stop training.

[0062] That is, each time a new electromyographic response is actually measured Then, the prediction model will be updated. The parameters are adjusted to adapt to the patient's real-time changes, and the updated formula is as follows: ; in, The model parameter vector is the prediction model. The coefficient; This is the gain matrix; This is the same regression vector as in the prediction model.

[0063] If the parameter vector is calculated Norm change rate If the norm change rate If the value remains below the norm threshold, the adaptive prediction model can be considered to have converged.

[0064] Through an online parameter update mechanism, the model can continuously adapt to changes in the patient's muscle response characteristics, ensuring that control remains effective.

[0065] This application achieves an intelligent rehabilitation closed loop of "CSI assessment → goal setting → adaptive training → reassessment" through the coordinated work of parameters, making core muscle training both precise and personalized, thereby effectively relieving spasticity after brain injury.

[0066] In summary, this application acquires surface electromyography (EMG) signals and position transition start times under multiple body positions; based on the position transition start times and the EMG signals, it obtains the average amplitude, maximum voluntary contraction amplitude, left-right symmetry index, stability index, and pre-activation time; based on the calculated average amplitude, maximum voluntary contraction amplitude, left-right symmetry index, stability index, and pre-activation time, it obtains a core muscle group function score; based on the core muscle group function score, left-right symmetry index, stability index, and pre-activation time, it determines the patient's training position, training goals, training mode, initial treatment parameters, and treatment optimization goals; based on the patient's training position, training goals, training mode, initial treatment parameters, and treatment optimization goals, it guides the patient in adaptive rehabilitation training, collects the patient's core muscle group EMG signals in real time, dynamically adjusts stimulation parameters through an adaptive prediction model and the EMG signals, and completes post-brain injury spasticity rehabilitation auxiliary training after the adaptive prediction model converges. This reduces reliance on the therapist's subjective experience and achieves precise rehabilitation training.

[0067] This application can dynamically adjust the stimulation mode and intensity based on the patient's real-time electromyographic response, improving training efficiency and increasing adaptability. This application can also formulate targeted training goals based on the individual patient's weaknesses (such as poor symmetry, insufficient stability, delayed activation, etc.).

[0068] For those consistent with the above, please refer to Figure 2 , Figure 2 This application provides a schematic diagram of the structure of a rehabilitation assistive training device. For example... Figure 2 As shown, it includes a peripheral intervention module and a central intervention module, and also includes: The data acquisition module 201 is used to acquire surface electromyography signals and the start time of body position transitions under multiple body position states; The index determination module 202 is used to obtain the average amplitude, maximum voluntary contraction amplitude, left-right symmetry index, stability index, and pre-activation time based on the body position change start time and the surface electromyography signal. The scoring determination module 203 is used to obtain a core muscle group function score based on the calculated average amplitude, the maximum voluntary contraction amplitude, the left-right symmetry index, the stability index, and the pre-activation time. The training parameter determination module 204 is used to determine the patient's training position, training goals, training mode, initial treatment parameters, and treatment optimization goals based on the core muscle group function score, left-right symmetry index, stability index, and pre-activation time. The training adjustment determination module 205 is used to guide the patient to conduct adaptive rehabilitation training based on the patient's training position, the patient's training goal, the patient's training mode, the initial treatment parameters, and the treatment optimization goal. It collects the surface electromyography (EMG) signals of the patient's core muscle groups in real time, dynamically adjusts the stimulation parameters through the adaptive prediction model and the surface EMG signals, and completes the auxiliary training for spasticity rehabilitation after brain injury after the adaptive prediction model converges.

[0069] like Figure 3 As shown, this application embodiment also provides a terminal device 2, which includes: at least one processor 20, a memory 21, and a computer program 22 stored in the memory 21 and executable on the at least one processor. The processor 20 and the memory 21 are connected. When the processor 20 executes the computer program 22, it implements the steps in the embodiments of the HIC value prediction method or the vehicle data augmentation method.

[0070] This application also provides a computer storage medium storing a computer program for electronic data exchange, which causes a computer to perform some or all of the steps of any of the brain injury post-spasticity rehabilitation auxiliary training methods described in the above method embodiments.

[0071] This application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program that causes a computer to perform some or all of the steps of any of the brain injury spasticity rehabilitation auxiliary training methods described in the above method embodiments.

[0072] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable storage medium can include at least: any entity or device capable of carrying computer program code to a device / terminal equipment, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable storage media cannot be electrical carrier signals or telecommunication signals.

[0073] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0074] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0075] In the embodiments provided in this application, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling or direct coupling or communication connection may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0076] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

Claims

1. A method for assisting in the rehabilitation training of spasticity after brain injury, characterized in that, The auxiliary training methods for post-brain injury spasticity rehabilitation include: Acquire surface electromyography signals and the start time of position transitions under multiple body positions; Based on the start time of the body position change and the surface electromyography signal, the average amplitude, maximum voluntary contraction amplitude, left-right symmetry index, stability index, and pre-activation time are obtained. The core muscle group function score is obtained based on the calculated average amplitude, the maximum voluntary contraction amplitude, the left-right symmetry index, the stability index, and the pre-activation time. Based on core muscle group function scores, left-right symmetry indicators, stability indicators, and pre-activation time, determine the patient's training position, training goals, training mode, initial treatment parameters, and treatment optimization goals. Based on the patient's training position, training goals, training mode, initial treatment parameters, and optimization goals, the patient is guided to undergo adaptive rehabilitation training. Surface electromyography (EMG) signals of the patient's core muscle groups are collected in real time. Stimulation parameters are dynamically adjusted using an adaptive prediction model and the EMG signals. After the adaptive prediction model converges, the patient completes auxiliary rehabilitation training for spasticity following brain injury.

2. The method for assisting in the rehabilitation training of spasticity after brain injury according to claim 1, characterized in that, The process of obtaining average amplitude, maximum voluntary contraction amplitude, left-right symmetry index, stability index, and pre-activation time based on the body position change start time and the surface electromyography signal includes: The surface electromyography signals within the stable body position segment were rectified using a full-wave rectifier to obtain the root mean square (RMS) data. The average amplitude is obtained based on the root mean square value data; For each body position, a maximum voluntary isometric contraction test was performed, and the maximum amplitude value in the root mean square value data was selected to obtain the maximum voluntary contraction amplitude. The amplitudes of the left and right muscle groups were filtered from the root mean square data, and the left-right symmetry index was obtained based on the amplitudes of the left and right muscle groups. Based on the root mean square value data and the average amplitude, a stability index is obtained; The pre-activation time is obtained based on the start time of body position change and the onset time of electromyographic activation in the surface electromyographic signal.

3. The method for assisting in the rehabilitation training of spasticity after brain injury according to claim 2, characterized in that, The step of obtaining the average amplitude based on the root mean square value data includes: ； in, Let J be the root mean square values ​​of the j windows; This represents the number of sliding windows within the stable segment.

4. The method for assisting in the rehabilitation training of spasticity after brain injury according to claim 2, characterized in that, The process of determining the variance of the amplitude signal within the stable segment from the root mean square (RMS) data to obtain a stability index includes: ； in, For stability indicators; This represents the total number of root mean square values ​​within the stable segment. Let be the root mean square value of the j-th window; The average amplitude.

5. The method for assisting in the rehabilitation training of spasticity after brain injury according to claim 1, characterized in that, The core muscle group function score is obtained based on the calculated average amplitude, the maximum voluntary contraction amplitude, the left-right symmetry index, the stability index, and the pre-activation time, including: ； in, Assess core muscle function. Weighting based on body position; Position the patient correctly; for Average amplitude under body position; for Maximum voluntary contraction amplitude under body position; The first evaluation coefficient; It is an indicator of left-right symmetry; This is the second evaluation coefficient; For stability indicators; The maximum acceptable variance; This is the third evaluation coefficient; This is the pre-activation time; To activate the reference time.

6. The method for assisting in the rehabilitation training of spasticity after brain injury according to claim 1, characterized in that, Based on core muscle function scores, bilateral symmetry indicators, stability indicators, and pre-activation time, the patient's training position, training goals, training mode, and initial treatment parameters are determined, including: Positions with low core muscle function scores were selected to determine the patient's training positions. Based on left-right symmetry indicators, stability indicators, and pre-activation time, the patient's training goals are obtained; Select the stimulation device corresponding to the muscle group that corresponds to the patient's training goal to obtain the patient's training mode; Initial treatment parameters and treatment optimization objectives are determined using a multi-objective optimization function.

7. The method for assisting in the rehabilitation training of spasticity after brain injury according to claim 1, characterized in that, The real-time acquisition of surface electromyography (EMG) signals of the patient's core muscle groups, the dynamic adjustment of stimulation parameters using an adaptive prediction model and the EMG signals, and the completion of post-brain injury spasticity rehabilitation auxiliary training after the adaptive prediction model converges include: In each control cycle k, the current electromyographic signal is acquired, and the real-time electromyographic amplitude is extracted. Predict future electromyographic responses using the current adaptive prediction model; The optimization problem is solved by rolling optimization of the objective function to determine the treatment parameters for the next time step; By updating the model parameters using recursive least squares with a forgetting factor, the model gradually adapts to the patient's individual response characteristics, and completes the auxiliary training for post-brain injury spasticity rehabilitation after the adaptive prediction model converges.

8. A rehabilitation assistive training device, characterized in that, It includes peripheral intervention modules and central intervention modules, and also includes: The data acquisition module is used to acquire surface electromyography signals and the start time of position transitions under multiple body position states; The index determination module is used to obtain the average amplitude, maximum voluntary contraction amplitude, left-right symmetry index, stability index, and pre-activation time based on the body position change start time and the surface electromyography signal. The scoring module is used to obtain a core muscle group function score based on the calculated average amplitude, the maximum voluntary contraction amplitude, the left-right symmetry index, the stability index, and the pre-activation time. The training parameter determination module is used to determine the patient's training position, training goals, training mode, initial treatment parameters, and treatment optimization goals based on core muscle group function scores, left-right symmetry indicators, stability indicators, and pre-activation time. The training adjustment and determination module is used to guide the patient to conduct adaptive rehabilitation training based on the patient's training position, the patient's training goal, the patient's training mode, the initial treatment parameters, and the treatment optimization goal. It collects the surface electromyography (EMG) signals of the patient's core muscle groups in real time, dynamically adjusts the stimulation parameters through an adaptive prediction model and the surface EMG signals, and completes the auxiliary rehabilitation training for spasticity after brain injury after the adaptive prediction model converges.

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the brain injury spasticity rehabilitation auxiliary training method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the brain injury spasticity rehabilitation auxiliary training method as described in any one of claims 1 to 7.

Images