Method and system for analyzing coupling between brain and muscle function changes based on two-way delay compensation

Abstract

The application discloses a brain-muscle function change coupling analysis method based on bidirectional delay compensation, which comprises the following steps: designing an experiment paradigm, collecting signals and pre-processing the signals; determining a bidirectional cortical muscle delay estimation algorithm based on a sliding window and transfer entropy TE, and quantifying the conduction time of the EEG->EMG outgoing path and the EMG->EEG incoming path of the cortical muscle of a test person; performing signal time alignment compensation, calculating the transfer entropy with delay compensation and the average delay compensation transfer entropy, and constructing a directional delay compensation directional cortical muscle connection CMC network based on the transfer entropy with delay compensation and the average delay compensation transfer entropy; extracting evaluation features, and performing brain-muscle function change coupling analysis based on the evaluation features; and the evaluation features are extracted based on the dimensions of the CMC coupling strength, the hemispheric lateralization index and the correlation of the delay-motor function behavior index score based on conduction delay, hemispheric division and task stage. The application also discloses a corresponding system, an electronic device and a computer readable storage medium.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of neuro-rehabilitation medicine and biomedical signal processing technology, and in particular to a brain-muscle function change coupling analysis method and system based on bidirectional delay compensation. BACKGROUND

[0002] There is a time delay (usually 20-30 ms) in the transmission of electroencephalogram (EEG) and electromyogram (EMG) signals. Traditional cortico-muscular coupling (CMC) analysis does not compensate for the delay, resulting in a time mismatch of the signals, which can mask the true functional coupling and reduce the sensitivity of the evaluation. In addition, existing delay estimation algorithms (such as the phase method and the optimization method) can only estimate the delay in a single direction (such as the outgoing direction), and cannot capture the delay changes in the afferent pathway, which is crucial for motor control.

[0003] To solve the above problems, researchers have tried to quantify the cortico-muscular functional connection through CMC analysis, but the existing technology still has two key bottlenecks: traditional CMC analysis mostly uses coherence methods, which can only capture linear interactions and cannot reflect the nonlinear neural interactions between the cortex and muscles (such as the complex regulatory relationships during the brain network reorganization process after stroke); neural signals during motor tasks are non-stationary (such as the differences in neural activity during the grip force rising period and the maintenance period), and traditional methods do not use dynamic analysis techniques such as sliding windows, so they cannot capture the CMC changes in different task stages

[0004] Therefore, there is an urgent need to design a CMC analysis method that can estimate the bidirectional delay of the cortex and muscles, compensate for the time mismatch of the signals, and capture dynamic nonlinear interactions, providing an objective and sensitive evaluation tool for brain-muscle function analysis and promoting its transition from "experience-driven" to "data-driven". SUMMARY

[0005] The purpose of the present application is to provide a method and system for analyzing the coupling of brain and muscle function changes based on bidirectional delay compensation, which is suitable for brain-computer interaction, motor function evaluation and other scenarios, and promotes the transformation from "experience-driven" to "data-driven". The method and system are based on the correlation mechanism of "cortical muscle bidirectional delay-function coupling-motor function", combined with transfer entropy (TE) and sliding window technology, to design a set of cortical muscle connection (CMC) analysis framework with bidirectional delay compensation. First, collect and preprocess the EEG and EMG signals, then propose a bidirectional brain-muscle conduction delay estimation algorithm, which globally optimizes the estimation of downlink and uplink conduction time by maximizing the significant transfer entropy unit across channels and time windows. Next, the estimated delay is compensated, and the high-sensitivity delay-compensated directional brain-muscle coupling strength is calculated. Finally, features are extracted from multiple dimensions such as conduction delay, coupling strength (hemisphere and task phase), hemisphere lateralization index, and delay-motor function behavior index score correlation, and the brain-muscle function changes are quantified through statistical analysis. Thus, the problems of subjectivity, low sensitivity, and inability to quantify cortical muscle bidirectional interaction in the prior art are solved.

[0006] The first aspect of the present application is to provide a method for analyzing the coupling of brain and muscle function changes based on bidirectional delay compensation, comprising:

[0007] S1, designing an experimental paradigm, based on which signal acquisition and signal preprocessing are performed;

[0008] S2, determining a bidirectional cortical muscle delay estimation algorithm based on sliding window and transfer entropy TE, and quantifying the conduction time of the EEG→EMG outgoing path and the EMG→EEG incoming path of the cortical muscle of the test subject based on the bidirectional cortical muscle delay estimation algorithm and the preprocessed signals;

[0009] S3, performing signal time alignment compensation based on the conduction time of the EEG→EMG outgoing path and the EMG→EEG incoming path, calculating the transfer entropy with delay compensation and the average delay compensation transfer entropy, and constructing a directional delay-compensated directional cortical muscle connection (CMC) network based on the transfer entropy with delay compensation and the average delay compensation transfer entropy, wherein the directional delay-compensated directional CMC network is used to visually display the strength distribution of the cortical muscle bidirectional functional connection;

[0010] S4, extracting evaluation features, and performing brain-muscle function change coupling analysis based on the evaluation features; wherein the evaluation features are extracted based on four dimensions of conduction delay, CMC coupling strength in hemispheres and task phases, hemisphere lateralization index and delay-motor function behavior index score correlation.

[0011] Preferably, S1 comprises:

[0012] S11, collecting electroencephalogram (EEG) signals, electromyogram (EMG) signals and behavioral data of a tester;

[0013] S12, preprocessing the EEG signals, EMG signals and behavioral data to remove noise and artifacts and reserve effective neural electrical signals.

[0014] Preferably, the preprocessing comprises:

[0015] (1) performing signal down-sampling, including: down-sampling the EEG signals from 2000 Hz to 1000 Hz, down-sampling the EMG signals from 2000 Hz to 1000 Hz, and down-sampling a grip strength signal in the behavioral data from 200 Hz to 1000 Hz;

[0016] (2) performing filtering processing on noise, including: extracting beta band signals by using a band-pass filter of 13-30 Hz; and removing power frequency noise by using a 50 Hz notch filter; wherein the beta band signals are a cortical muscle function connection dominant frequency band;

[0017] (3) removing artifacts, including: identifying and removing ocular artifacts, electrocardiogram artifacts and EMG interference artifacts in the EEG signals based on independent component analysis (ICA) combined with an ICLABEL plug-in;

[0018] (4) performing signal re-referencing and segmentation: re-referencing the EEG signals to an average reference of all channels; dividing a grip tracking period into a resting stage S0, a ramp-up stage S1 and a maintenance stage S2 according to a profile of the grip strength signal, and synchronously segmenting the EEG signals and the EMG signals according to the segmentation of the grip tracking period;

[0019] (5) performing baseline correction: taking the stage signal of the resting stage S0 as a baseline, and performing baseline correction on the stage signals of the ramp-up stage S1 and the maintenance stage S2.

[0020] Preferably, the S2 comprises:

[0021] S21, setting a sliding window, including: dividing the EEG signals and the EMG signals after the synchronous segmentation by using an overlapping sliding window with a length of 500 ms and a step length of 250 ms;

[0022] S22, performing phase space reconstruction, including: performing phase space reconstruction on the EEG signals and the EMG signals in each sliding window to obtain formula (1) and formula (2): (1);

[0023] (2);

[0024] In the formula, This represents the time series of EEG signals. This represents the time series of electromyographic (EMG) signals. The associated time delay is determined by the correlation integral method. , All represent the embedding dimension, determined by the Cao criterion;

[0025] S23, Calculate the transfer entropy TE, including: calculating the transfer entropy in the outgoing direction EEG→EMG respectively. The transfer entropy with the input direction EMG→EEG ,in Characterizing the predictive ability of EEG signals to EMG signals. Transition entropy characterizes the predictive ability of electromyography (EMG) signals on electroencephalography (EEG) signals. and transfer entropy The calculation formulas are shown in (3) and (4) respectively:

[0026] (3);

[0027] (4);

[0028] In the formula, H(·) represents the conditional entropy, and ε represents the time interval. In this embodiment, it is set to 1. Those skilled in the art can make appropriate settings as needed, all of which are within the protection scope of this invention. They are respectively Embedding vectors of EEG and EMG signals at any given time; and Both are scalar random variables, representing two time series, EEG signals and EMG signals, respectively, at a specific future moment. The observed values ​​or state, Indicates from the current reference time Beginning, moving towards the future A moment after that time;

[0029] S24, Calculate the transfer entropy (DTE) with delay compensation, including: candidate delay based on the output direction. Time-align the signals to calculate the delay-compensated transfer entropy from EEG to EMG in the outgoing direction. And candidate delays based on the incoming direction Time-align the signals to calculate the delay-compensated transfer entropy from EMG to EEG in the incoming direction. The transfer entropy of the outgoing direction EEG→EMG with delay compensation Transfer entropy with delay compensation in the direction of incoming EMG→EEG The calculation formulas are shown in (5) and (6) respectively:

[0030] (5);

[0031] (6);

[0032] In the formula, is a sliding window index, and are channel indexes of electroencephalogram (EEG) signals and electromyogram (EMG) signals respectively; is a delay delayed electromyogram (EMG) signal embedding vector, is a delay delayed electroencephalogram (EEG) signal embedding vector; is a scalar random variable, representing the instantaneous amplitude or state of the signal from the electromyogram channel at time point in the first sliding analysis window ; is a scalar random variable, representing the instantaneous amplitude or state of the signal from the electroencephalogram channel at time point in the first sliding analysis window ;

[0033] S25, verifying the statistical significance of the delay-compensated transfer entropy DTE based on the surrogate data method, including: generating 10,000 phase randomized original EEG-EMG signals to obtain the surrogate data, calculating the delay-compensated transfer entropy DTE value of the surrogate data , and determining whether the delay-compensated transfer entropy DTE value of the original EEG-EMG signals is significantly higher than the delay-compensated transfer entropy DTE value of the surrogate data by t-test , if the delay-compensated transfer entropy DTE value of the original EEG-EMG signals is significantly higher than the delay-compensated transfer entropy DTE value of the surrogate data , it indicates that there is a directional causal interaction between the channel pairs;

[0034] S26, performing bidirectional delay optimization, including: designing an optimization criterion to maximize the significant DTE sum across EEG-EMG channel pairs and sliding windows, searching for an optimized outgoing direction optimal delay and an optimized incoming direction optimal delay , the optimized outgoing direction optimal delay and the optimized incoming direction optimal delay The optimization formula of S3 is shown in equation (7) and equation (8):

[0035] (7);

[0036] (8);

[0037] wherein, and are the channel index of electroencephalogram (EEG) signal and electromyogram (EMG) signal respectively, is the sliding window index, is the DTE change tracking function, is the indicator function, taking 1 when the delay-compensated transfer entropy (DTE) is significant, otherwise taking 0, , are the estimated efferent and afferent corticomuscular conduction time respectively; is a positive integer, indicating the upper limit of the summation symbol , representing the total number of EEG signal channels in the system, and the channel index traverses from 1 to , covering all available EEG signal channels; is a positive integer, indicating the upper limit of the summation symbol , representing the total number of EMG signal channels in the system, and the channel index traverses from 1 to , covering all available EMG signal channels; is a positive integer, indicating the upper limit of the summation symbol , representing the total number of sliding windows divided over the entire data time series, and the sliding window index traverses from 1 to , covering all analysis time segments from the beginning to the end of the experiment.

[0038] Preferably, the S3 comprises:

[0039] S31, time-aligning the EEG signal and EMG signal based on the optimized efferent optimal delay and the optimized afferent optimal delay ;

[0040] S32, substituting the time-aligned EEG signal and EMG signal into equation (9) and (10) to calculate the delay-compensated efferent DTE with delay compensation , and the delay-compensated afferent DTE with delay compensation ;

[0041] (9);

[0042] (10);

[0043] wherein, is the delayed EMG embedding vector, is the delayed EEG embedding vector; and ε represents the time interval; represents the instantaneous amplitude or state of the electromyography (EMG) signal at a future time ; represents the instantaneous amplitude or state of the electroencephalogram (EEG) signal at a future time ; wherein: is the current analysis time, i.e., the reference point; and are the estimated optimal physiological conduction delays obtained by global search through the aforementioned optimization formulas (7) and (8), this composite time point represents the exact future time at which the result signal of interest, either electromyography or electroencephalogram, should appear after compensating for the estimated, pathway-specific physiological conduction time;

[0044] S33, calculating the average delay-compensated transfer entropy, comprising: verifying the statistical significance of the delay-compensated transfer entropy under the same hemisphere and the same task stage based on the surrogate data method to obtain all significant delay-compensated transfer entropies; and averaging all significant delay-compensated transfer entropies to obtain the average delay-compensated transfer entropy , which is used as a quantitative indicator of the CMC coupling strength to represent the cortical muscle function coupling strength, and the calculation formula is shown in formula (11):

[0045] (11);

[0046] wherein, are the selected EEG channel set, EMG channel set and sliding window set, respectively, and N(·) is the set cardinality, is the optimal delay of the corresponding direction;

[0047] S34, constructing a visualized directional delay-compensated directional cortical muscle connection (CMC) network, comprising: taking the EEG channels and the EMG channels as nodes, and taking the average delay-compensated transfer entropy as the edge weight, to construct the directional delay-compensated directional cortical muscle connection (CMC) network, wherein the EEG channels are sensor motor cortex regions. ​

[0048] Preferably, 18 channels of the sensor motor cortex region are selected as the EEG channels, the 18 channels include FC1, FC3, FC5, C1, C3, C5, CP1, CP3, CP5 channels of the left hemisphere and FC2, FC4, FC6, C2, C4, C6, CP2, CP4, CP6 channels of the right hemisphere; according to the target functional side of the tester, the 18 channels are classified into target side hemisphere channels and control side hemisphere channels respectively.

[0049] Preferably, the S4 comprises:

[0050] S41, extracting a cortical muscle conduction time feature based on the conduction delay dimension, comprising: extracting an optimized efferent direction optimal delay and an optimized afferent direction optimal delay as the cortical muscle conduction time feature, so as to reflect the conduction efficiency of the neural pathway;

[0051] S42, extracting a CMC coupling strength feature based on the CMC coupling strength dimension of the hemispheric division and the task phase division, comprising: extracting average delay compensation transfer entropy of the target side hemisphere and the control side hemisphere in the resting phase S0, the ramp-up phase S1 and the maintenance phase S2 in the efferent and afferent directions respectively as the CMC coupling strength feature, so as to reflect the interaction strength of the cortical muscle function;

[0052] S43, extracting the hemispheric lateralization index based on the hemispheric lateralization index dimension, comprising: calculating the hemispheric lateralization index based on the average delay compensation transfer entropy of the target side and the control side , so as to quantify the degree of lateralization of motor control function, the hemispheric lateralization index The calculation formula of the hemispheric lateralization index

[0053] is shown in formula (12): (12);

[0054] In the formula, is the average delay compensation transfer entropy of the target side hemisphere, is the average delay compensation transfer entropy of the control side hemisphere, the value range of the hemispheric lateralization index

[0055] S44, extracting a behavior and motor function score feature based on the delay-motor function behavior index score correlation dimension, comprising: taking the RMSE of the grip tracking task as a behavior feature of fine motor control ability evaluation, and taking the behavior and motor function score as the motor function score feature.

[0056] S45, performing brain-muscle function change coupling analysis based on the evaluation features.

[0057] The second aspect of the present application provides a brain-muscle function change coupling analysis system based on bidirectional delay compensation, which is used to implement the method of the first aspect, and comprises:

[0058] A data acquisition module (101) is configured to design an experimental paradigm, and perform signal acquisition and signal preprocessing based on the experimental paradigm.

[0059] A conduction time optimization module (102) is configured to determine a bidirectional cortical muscle delay estimation algorithm based on a sliding window and a transfer entropy TE, and quantify conduction times of an EEG→EMG efferent pathway and an EMG→EEG afferent pathway of a test subject's cortical muscle based on the bidirectional cortical muscle delay estimation algorithm and the preprocessed signal.

[0060] A network construction and coupling strength delay compensation module (103) is configured to perform signal time alignment compensation based on the conduction times of the EEG→EMG efferent pathway and the EMG→EEEG afferent pathway, calculate a delay-compensated transfer entropy and an average delay-compensated transfer entropy, and construct a directional delay-compensated directional cortical muscle connection (CMC) network based on the delay-compensated transfer entropy and the average delay-compensated transfer entropy, wherein the directional delay-compensated directional CMC network is used to visually display the strength distribution of the bidirectional functional connection of the cortical muscle.

[0061] A brain-muscle function change coupling analysis module (104) is configured to extract evaluation features, and perform brain-muscle function change coupling analysis based on the evaluation features, wherein the evaluation features are extracted based on four dimensions of conduction delay, CMC coupling strength of hemispheric division and task stage division, hemispheric lateralization index, and correlation of delay-motor function behavior index score.

[0062] The third aspect of the present application provides an electronic device comprising a processor and a memory, wherein the memory stores a plurality of instructions, and the processor is configured to read the instructions and perform the method of the first aspect.

[0063] The fourth aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a plurality of instructions, and the plurality of instructions can be read and executed by a processor to perform the method of the first aspect.

[0064] The method and system of the present application have the following advantages:

[0065] 1. The present application realizes accurate quantification and bidirectional separation of cortical muscle conduction time, and breaks through the limitation of one-way evaluation. ​

[0066] Traditional methods are often limited to assessing one-way conduction from brain to muscle (EEG→EMG). The present invention can accurately quantify the conduction time of both outgoing (EEG→EMG) and incoming (EMG→EEG) pathways by introducing a two-way cortical muscle delay estimation algorithm. This is the first time that the separation and independent measurement of the "up" and "down" signal conduction speeds in the cortical muscle loop are achieved in method, providing key data for in-depth study of the neural mechanisms of motor control and sensory feedback, and breaking through the limitations of traditional one-way evaluation.

[0067] 2、Construct a high-fidelity directional functional connectivity network, significantly improve the authenticity and intuitiveness of brain-muscle coupling analysis

[0068] The present invention innovatively introduces a delay compensation mechanism to solve the calculation error of coupling strength caused by asynchronous signal conduction time. By calculating the transfer entropy with delay compensation and constructing a directional delay-compensated CMC network, the strength of the two-way functional connection between the brain and the muscle can be more realistically reflected. This network visually displays the spatial distribution of coupling strength in an intuitive and visual manner, allowing researchers to easily identify the core brain-muscle interaction pathways, greatly improving the accuracy and interpretability of the analysis results.

[0069] 3、Provide a multi-dimensional and comprehensive evaluation system to realize panoramic correlation analysis from neurophysiology to behavior performance

[0070] The system extracts comprehensive evaluation features through four dimensions (conduction delay, hemispheric / phase-specific CMC strength, hemispheric lateralization index, and delay-behavioral score correlation). This multi-dimensional framework avoids the one-sidedness of a single indicator and can reveal the time characteristics (conduction speed), spatial characteristics (left and right hemispheric division and lateralization), state characteristics (coupling dynamics in different task stages), and functional significance (neural conduction efficiency and correlation with motor behavior performance), thereby achieving a panoramic and deep-level interpretation of brain-muscle functional coupling changes.

[0071] 4、Provide a standardized and locatable objective quantification tool for motor function evaluation and neural rehabilitation research

[0072] The present invention integrates experimental paradigm design, signal acquisition, preprocessing, core algorithm calculation, and comprehensive analysis into a complete system. This modular and process-oriented design makes the entire evaluation process highly standardized, reduces human error, and ensures the repeatability and comparability of the results. By precisely locating the abnormal pathways of conduction delay (outgoing or incoming problem) and the brain regions with weakened coupling, the present invention can provide strong data support for precise evaluation of motor dysfunction, objective monitoring of rehabilitation efficacy, and development of individualized rehabilitation programs in clinical practice.

[0073] 5、Through the two-way delay compensation technology, based on the accurate estimation and compensation of the two-way conduction time of EEG and EMG, the coupling strength is calculated more reliably, the delay existing in the brain-muscle path is reduced, and the two-way information flow delay between the brain and the muscle is accurately quantified to calculate the coupling strength more reliably. BRIEF DESCRIPTION OF DRAWINGS

[0074] In order to more clearly illustrate the technical solutions in the specific embodiments or related art of the present application, the drawings needed in the specific embodiments or related art description will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application, and all other embodiments obtained by those skilled in the art without creative labor based on these drawings also belong to the protection scope of the present application.

[0075] Figure 1 The flow chart of the brain-muscle function change coupling analysis method based on two-way delay compensation provided according to the embodiment of the present application is provided.

[0076] Figure 2 The flow chart of step S1 of the brain-muscle function change coupling analysis method based on two-way delay compensation provided according to the embodiment of the present application is provided.

[0077] Figure 3 The flow chart of step S2 of the brain-muscle function change coupling analysis method based on two-way delay compensation provided according to the embodiment of the present application is provided.

[0078] Figure 4 The flow chart of step S3 of the brain-muscle function change coupling analysis method based on two-way delay compensation provided according to the embodiment of the present application is provided.

[0079] Figure 5 The flow chart of step S4 of the brain-muscle function change coupling analysis method based on two-way delay compensation provided according to the embodiment of the present application is provided.

[0080] Figure 6 The system architecture diagram of the flow chart of step S1 of the brain-muscle function change coupling analysis method based on two-way delay compensation provided according to the embodiment of the present application is provided.

[0081] Figure 7 The structure diagram of the electronic device provided according to the embodiment of the present application is provided. DETAILED DESCRIPTION

[0082] The technical solutions of the present application will be described below in conjunction with the drawings. Obviously, the described embodiments are some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor also belong to the protection scope of the present application.

[0083] In the description of the present application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer" and the like indicate the orientation or positional relationship shown in the drawings, which are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application. In addition, the terms "first", "second", "third" are only for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0084] In the description of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms "mounting", "connection", "connection" should be understood in a broad sense, for example, it can be fixed connection, or detachable connection, or integrally connected; it can be mechanical connection, or electrical connection; it can be directly connected, or indirectly connected through intermediate medium, or it can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0085] The following embodiments aim to solve the following core technical problems:

[0086] (1) Realize the cooperative optimization of multiple performance indicators: design a control method that can not only ensure high-precision trajectory tracking, but also strictly constrain and manage multiple performance indicators such as overshoot and convergence time, and comprehensively improve the efficiency and quality of rehabilitation training.

[0087] (2) Give dynamic adaptive ability to performance boundary: solve the safety problem that fixed performance boundary may cause in sudden situation, design a soft performance boundary that can relax or tighten dynamically according to system real-time, balance optimal performance under the premise of safety.

[0088] (3) Establish the association of disturbance and constraint and intelligently adjust the boundary: clearly propose the concept of "safety boundary" and build its linkage mechanism with performance boundary. When the system detects that the tracking error exceeds the safety tolerance due to disturbance, it can automatically trigger the adjustment strategy of temporarily relaxing the performance boundary, thereby resolving the conflict and ensuring the safe and stable operation of the system.

[0089] (4) Efficiently fuse intelligent approximation and constraint learning: use the Actor-Critic framework of reinforcement learning and the approximation ability of neural network to efficiently estimate and compensate for the lumped disturbance in the system; and integrate the preset performance constraint into the learning framework to significantly reduce the policy search space, accelerate the convergence speed of the control strategy, and improve the learning efficiency.

[0090] The professional terms and their meanings involved in the present embodiment are as follows:

[0091] (1) Aggregate disturbance: refers to the summation of all uncertainties acting on the controlled plant. In the context of this patent, it is a unified concept, which specifically includes the un-modeled dynamics, parameter variations, etc. of the system, as well as the external disturbances to which the system is subjected.

[0092] (2) Soft Prescribed Performance: a dynamic performance constraint mechanism. It specifies the transient and steady-state behavior of the system tracking error (such as convergence speed, overshoot, and steady-state error band) through a time-varying, state-dependent performance boundary function.

[0093] (3) Critic-Actor based Reinforcement Learning Framework: an adaptive optimization structure that employs two neural networks to learn in parallel.

[0094] (4) Command Filtering Technique: a numerical technique used in the design framework of backstepping to generate the virtual control law and its derivative. It takes the virtual control law calculated in the previous design step as the command input to a filter (usually a first or second order linear filter), and the output of the filter is the smoothed virtual control command, while its analytical derivative can be directly obtained. The main purpose of this technique is to fundamentally avoid the "analytical differentiation explosion" problem in backstepping, greatly simplifying the controller structure, and introducing amplitude and rate limits by setting filter parameters.

[0095] (5) Filtered Error Compensation Signal: a dynamic signal specially designed to offset the phase lag and amplitude attenuation introduced by the command filter dynamics.

[0096] Embodiment One

[0097] As shown in the embodiment, a brain-muscle function change coupling analysis method based on bidirectional delay compensation is provided, which comprises: Figure 1 S1, designing an experimental paradigm, based on which signal acquisition and signal preprocessing are performed;

[0098] As shown in the preferred embodiment, S1 comprises:

[0099] Figure 2 S11, acquiring electroencephalogram (EEG) signals, electromyogram (EMG) signals, and behavioral data of the test personnel;

[0100] S11, acquiring electroencephalogram (EEG) signals, electromyogram (EMG) signals, and behavioral data of the test personnel;

[0101] ​In this embodiment, the collection of electroencephalogram (EEG) signals adopts a 64-channel Neuroscan SynAmps system, which follows the international 10-10 electrode system, and the electrode impedance is controlled below 10 kΩ; the collection of electromyogram (EMG) signals adopts a Trigno Wireless EMG system, and the EMG electrodes of the tester are pasted on the corresponding muscle area of the target side; the behavioral data, specifically the grip strength signal, is collected by a Vernier Hand Dynamometer (Vernier handgrip), with a sampling rate of 200 Hz, for dividing the task stages and calculating the root mean square error (RMSE).

[0102] S12, preprocessing the electroencephalogram (EEG) signals, electromyogram (EMG) signals and behavioral data to remove noise and artifacts and retain effective neural electrical signals;

[0103] In this embodiment, the preprocessing method comprises:

[0104] (1) signal downsampling, including: downsampling the electroencephalogram (EEG) signals from 2000 Hz to 1000 Hz, downsampling the electromyogram (EMG) signals from 2000 Hz to 1000 Hz, and downsampling the grip strength signal in the behavioral data from 200 Hz to 1000 Hz;

[0105] (2) noise filtering, including: extracting beta band signals by using a 13-30 Hz band-pass filter; removing power frequency noise by using a 50 Hz notch filter; wherein the beta band signals are the dominant frequency band of cortical muscle functional connectivity;

[0106] (3) artifact removal, including: identifying and removing electrooculogram (EOG) artifacts, electrocardiogram (ECG) artifacts and electromyogram (EMG) interference artifacts in the electroencephalogram (EEG) signals based on independent component analysis (ICA) combined with ICLABEL plug-in;

[0107] (4) signal re-referencing and segmentation: re-referencing the electroencephalogram (EEG) signals to the average reference of all channels; dividing the grip strength tracking period into a resting stage S0, a ramp-up stage S1 and a maintenance stage S2 according to the profile of the grip strength signal, and synchronously segmenting the electroencephalogram (EEG) signals and the electromyogram (EMG) signals according to the segmentation of the grip strength tracking period;

[0108] (5) baseline correction: taking the stage signal of the resting stage S0 as the baseline, and performing baseline correction on the stage signals of the ramp-up stage S1 and the maintenance stage S2.

[0109] S2, a bidirectional cortical muscle delay estimation algorithm is determined based on the sliding window and the transfer entropy TE, and the conduction time of the EEG→EMG efferent pathway and the EMG→EEG afferent pathway of the cortical muscle of the test personnel is quantified based on the bidirectional cortical muscle delay estimation algorithm and the preprocessed signal.

[0110] like Figure 3 As shown, in a preferred embodiment, S2 includes:

[0111] S21, Set a sliding window, including: using an overlapping sliding window with a length of 500ms and a step size of 250ms to divide the EEG signal and EMG signal after the synchronous segmentation;

[0112] S22, perform phase space reconstruction, including: perform phase space reconstruction on the EEG signal and EMG signal in each sliding window to obtain equation (1) and equation (2):

[0113] (1);

[0114] (2);

[0115] In the formula, This represents the time series of EEG signals. This represents the time series of electromyographic (EMG) signals, where τ represents the associated time delay, determined by the correlation integral method. , All represent the embedding dimension, determined by the Cao criterion;

[0116] S23, Calculate the transfer entropy TE, including: calculating the transfer entropy in the outgoing direction EEG→EMG respectively. The transfer entropy with the input direction EMG→EEG ,in Characterizing the predictive ability of EEG signals to EMG signals. Transition entropy characterizes the predictive ability of electromyography (EMG) signals on electroencephalography (EEG) signals. and transfer entropy The calculation formulas are shown in (3) and (4) respectively:

[0117] (3);

[0118] (4);

[0119] In the formula, H(·) represents the conditional entropy, and ε represents the time interval. In this embodiment, it is set to 1. Those skilled in the art can make appropriate settings as needed, all of which are within the protection scope of this invention. They are respectively embedding vectors of electroencephalogram (EEG) and electromyogram (EMG) signals at time instant t; are scalar random variables representing the observation or state of the two time series of electroencephalogram (EEG) and electromyogram (EMG) signals at a certain future time instant t+T; denotes a time instant T ahead of the current reference time instant t;

[0120] S24, calculating the transfer entropy DTE with delay compensation, including: based on the candidate delay of the outgoing direction time aligning the signals, thereby calculating the transfer entropy of the outgoing direction EEG→EMG with delay compensation time aligning the signals, thereby calculating the transfer entropy of the incoming direction EMG→EEG with delay compensation The calculation formulas of the transfer entropy of the outgoing direction EEG→EMG with delay compensation

[0121] (5);

[0122] (6);

[0123] wherein, is the sliding window index, and are the channel indices of the electroencephalogram (EEG) and electromyogram (EMG) signals, respectively; is the embedding vector of the electromyogram (EMG) signal after delay T, is the embedding vector of the electroencephalogram (EEG) signal after delay T; is a scalar random variable representing the instantaneous amplitude or state of the signal from the electromyogram channel at time point t+T within the th sliding analysis window; is a scalar random variable representing the instantaneous amplitude or state of the signal from the electroencephalogram channel at time point t+T within the th sliding analysis window;

[0124] ​​​​​​​​​​​S25, verifying statistical significance of the delay-compensated transfer entropy DTE based on surrogate data method, including: generating the surrogate data by 10000 times of phase randomization on the original EEG-EMG signal, calculating the delay-compensated transfer entropy DTE value of the surrogate data , judging whether the delay-compensated transfer entropy DTE value of the original EEG-EMG signal is significantly higher than the delay-compensated transfer entropy DTE value of the surrogate data by t-test (p<0.05) , if the delay-compensated transfer entropy DTE value of the original EEG-EMG signal is significantly higher than the delay-compensated transfer entropy DTE value of the surrogate data , it indicates that there is directional causal interaction between the channel pairs;

[0125] S26, performing bidirectional delay optimization, including: designing optimization criteria to maximize the significant DTE sum of the EEG-EMG channel pairs in the sliding window, searching for the optimized out-directional optimal delay and the optimized in-directional optimal delay , the optimization formula of the optimized out-directional optimal delay and the optimized in-directional optimal delay is shown in formula (7) and formula (8) as follows:

[0126] (7);

[0127] (8);

[0128] In the formula, and are channel indexes of electroencephalogram EEG signal and electromyogram EMG signal respectively, is a sliding window index, is a DTE change tracking function, is an indicator function, taking 1 when the delay-compensated transfer entropy DTE is significant, otherwise taking 0, , are estimated out-directional and in-directional cortical muscle conduction times respectively; is a positive integer, indicating the upper limit of the summation symbol , representing the total number of electroencephalogram EEG signal channels in the system, and the channel index traverses from 1 to , covering all available electroencephalogram EEG signal channels; is a positive integer, indicating the upper limit of the summation symbol , representing the total number of electromyogram EMG signal channels in the system, and the channel index traverses from 1 to covers all available EMG channels; is a positive integer, representing the summation symbol upper limit, representing the total number of sliding windows divided on the entire data time series, sliding window index traverses from 1 to covers all analysis time segments from the beginning to the end of the experiment.

[0129] S3, based on the conduction time of the EEG→EMG efferent pathway and the EMG→EEG afferent pathway, compensates for signal time alignment, calculates delay-compensated transfer entropy and average delay-compensated transfer entropy, and constructs a directional delay-compensated directional corticomuscular connectivity (CMC) network based on the delay-compensated transfer entropy and the average delay-compensated transfer entropy, wherein the directional delay-compensated directional CMC network is used to visually display the strength distribution of the bidirectional functional connectivity of the cortex and muscles.

[0130] As shown in Figure 4 , as a preferred embodiment, S3 includes:

[0131] S31, based on the optimized efferent direction optimal delay and the optimized afferent direction optimal delay , time-aligns the electroencephalogram (EEG) signal and the electromyogram (EMG) signal;

[0132] S32, substitutes the time-aligned electroencephalogram (EEG) signal and the electromyogram (EMG) signal into formulas (9) and (10) to calculate the delay-compensated efferent direction EEG→EMG transfer entropy and the delay-compensated afferent direction EMG→EEG transfer entropy ;

[0133] (9);

[0134] (10);

[0135] In the formula, is the delay compensated EMG embedding vector, is the delay compensated EEG embedding vector; ε represents the time interval, which is set to 1 in this embodiment, and can be appropriately set by those skilled in the art as needed, all within the protection scope of the present application; represents the instantaneous amplitude or state of the electromyogram (EMG) signal at a future time ; represents the instantaneous amplitude or state of the electroencephalogram (EEG) signal at a future time the instantaneous amplitude or state of the signal; wherein: is the current analysis time, i.e. the reference point; and is the estimated optimal physiological conduction delay obtained by global search through the aforementioned optimization equations (7) and (8), This composite time point indicates the exact future time at which the result signal of interest, i.e. the EMG or EEG, should appear after compensating for the estimated, pathway-specific physiological conduction time;

[0136] S33, calculating the average delay-compensated transfer entropy, comprising: verifying the statistical significance of the delay-compensated transfer entropy in the same hemisphere and the same task phase based on the surrogate data method to obtain all significant delay-compensated transfer entropies; and averaging all significant delay-compensated transfer entropies to obtain the average delay-compensated transfer entropy , the average delay-compensated transfer entropy is used as a quantitative indicator of the CMC coupling strength to characterize the cortical muscle function coupling strength, and the calculation formula is shown in equation (11):

[0137] (11);

[0138] In the formula, are the selected EEG channel set, EMG channel set and sliding window set respectively, N(·) is the set cardinality, is the optimal delay of the corresponding direction;

[0139] S34, constructing a visualized directional delay-compensated directional cortical muscle connection CMC network, comprising: taking the EEG channels and the EMG channels as nodes, and taking the average delay-compensated transfer entropy as the edge weight, to construct a directional delay-compensated directional cortical muscle connection CMC network, wherein the EEG channels are sensor motor cortex regions.

[0140] As a preferred embodiment, 18 channels of the sensor motor cortex region are selected as the EEG channels, and the 18 channels include FC1, FC3, FC5, C1, C3, C5, CP1, CP3, CP5 channels of the left hemisphere and FC2, FC4, FC6, C2, C4, C6, CP2, CP4, CP6 channels of the right hemisphere.

[0141] As a preferred embodiment, according to the target functional side of the test personnel, the 18 channels are classified into target side hemisphere channels and control side hemisphere channels respectively, and are only used for calculation of the cortical muscle delay and the CMC coupling strength.

[0142] S4, extracting evaluation features and performing brain-muscle function change coupling analysis based on the evaluation features; wherein the evaluation features are extracted based on four dimensions of conduction delay, CMC coupling strength of split-hemisphere and split-task stage, hemispheric lateralization index, and correlation between delay-motor function behavioral index score .

[0143] As shown in Figure 5 , as a preferred embodiment, the S4 comprises:

[0144] S41, extracting cortical muscle conduction time features based on the conduction delay dimension, comprising: extracting optimized efferent direction optimal delay and optimized afferent direction optimal delay as the cortical muscle conduction time features, so as to reflect the conduction efficiency of neural pathways;

[0145] S42, extracting CMC coupling strength features based on the CMC coupling strength dimension of split-hemisphere and split-task stage, comprising: extracting average delay-compensated transfer entropy of efferent and afferent directions of target side hemisphere and control side hemisphere in resting stage S0, ramp-up stage S1 and maintenance stage S2 as the CMC coupling strength features, so as to reflect the interaction strength of cortical muscle function;

[0146] S43, extracting the hemispheric lateralization index feature based on the hemispheric lateralization index dimension, comprising: calculating the hemispheric lateralization index based on average delay-compensated transfer entropy of target side and control side, so as to quantify the lateralization degree of motor control function, the calculation formula of the hemispheric lateralization index is shown in formula (12):

[0147] (12);

[0148] wherein, is average delay-compensated transfer entropy of target side hemisphere, is average delay-compensated transfer entropy of control side hemisphere, the value range of the hemispheric lateralization index

[0149] S44, extracting behavioral and motor function score features based on the delay-motor function behavioral index score correlation dimension, comprising: taking RMSE of grip tracking task as behavioral feature of fine motor control ability evaluation, and taking behavioral and motor function score as the motor function score feature;

[0150] S45, Perform a coupling analysis of brain muscle function changes based on the assessment features.

[0151] Example 2

[0152] like Figure 6 As shown, this embodiment provides a brain-muscle function change coupling analysis system based on bidirectional delay compensation, used to implement the method of Embodiment 1, including:

[0153] Data acquisition module 101 is used to design experimental paradigms and perform signal acquisition and signal preprocessing based on the experimental paradigms;

[0154] The conduction time optimization module 102 is used to determine the bidirectional cortical muscle delay estimation algorithm based on the sliding window and the transfer entropy TE, and to quantify the conduction time of the EEG→EMG efferent pathway and the EMG→EEG infferent pathway of the cortical muscle of the test personnel based on the bidirectional cortical muscle delay estimation algorithm and the preprocessed signal.

[0155] The network construction and coupling strength delay compensation module 103 is used to perform signal time alignment compensation based on the transmission time of the EEG→EMG outgoing path and the EMG→EEG incoming path, calculate the delay-compensated transfer entropy and the average delay-compensated transfer entropy, and construct a directional delay-compensated directional corticomuscular connection CMC network based on the delay-compensated transfer entropy and the average delay-compensated transfer entropy. The directional delay-compensated directional corticomuscular connection CMC network is used to visually display the strength distribution of bidirectional functional connections between corticomuscular muscles.

[0156] The brain-muscle function change coupling analysis module 104 is used to extract assessment features and perform brain-muscle function change coupling analysis based on the assessment features; wherein, the assessment features are based on conduction delay, CMC coupling strength in hemispheric and task-specific stages, and hemispheric lateralization index. The data was extracted from four dimensions: the correlation between delayed-motion function behavioral index scores and other data.

[0157] Application scenarios:

[0158] 1. Motor function status assessment scenario: It is suitable for quantitative assessment of the cortical muscle functional connectivity status of healthy people and people with different motor functions. By analyzing characteristics such as efferent / apergent pathway conduction time, target side and control side DTE, lateralization index (LI), etc., it can accurately determine the cortical muscle interaction efficiency and motor control function level of the assessed object.

[0159] 2. Neurological function monitoring scenario: It can be integrated into a neurological function monitoring system to track the dynamic changes in brain-muscle coupling of the monitored subject in different activity states (such as resting and exercise training) in real time, providing data support for the early identification of cortical muscle dysfunction;

[0160] 3. The man-machine interaction system calibration scenario: in the research and application of man-machine interaction device (such as prosthesis control, motion auxiliary device) based on electroencephalogram-electromyogram signal, the brain-muscle conduction delay can be accurately estimated, the brain-muscle coupling signal processing algorithm can be optimized, and the response accuracy and collaborative stability of man-machine interaction can be improved.

[0161] The application further provides a memory which stores a plurality of instructions for implementing the method of embodiment one.

[0162] As shown in Figure 7 The application further provides an electronic device which comprises a processor 301 and a memory 302 connected with the processor 301, and the memory 302 stores a plurality of instructions which can be loaded and executed by the processor to enable the processor to execute the method of embodiment two and embodiment three.

[0163] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the application, and not to limit them; although the application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the application.

Claims

1. A method for coupled analysis of brain muscle function changes based on bidirectional delay compensation, characterized in that, include: S1. Design an experimental paradigm, and perform signal acquisition and signal preprocessing based on the experimental paradigm; S2, a bidirectional cortical muscle delay estimation algorithm is determined based on the sliding window and the transfer entropy TE, and the conduction time of the EEG→EMG efferent pathway and the EMG→EEG afferent pathway of the cortical muscle of the test personnel is quantified based on the bidirectional cortical muscle delay estimation algorithm and the preprocessed signal. S3, based on the conduction time of the EEG→EMG outgoing pathway and the EMG→EEG incoming pathway, signal time alignment compensation is performed; the delay-compensated transfer entropy and the average delay-compensated transfer entropy are calculated; and a directional delay-compensated directional corticomuscular connectivity (CMC) network is constructed based on the delay-compensated transfer entropy and the average delay-compensated transfer entropy. The directional delay-compensated directional corticomuscular connectivity CMC network is used to visually demonstrate the strength distribution of bidirectional functional connections between the cortex and muscles. S3 includes: S31, based on the optimized outgoing direction optimal delay And optimized input direction optimal delay The EEG and EMG signals were time-aligned. S32, Substitute the EEG and EMG signals aligned over the specified time into formulas (9) and (10) to calculate the delay-compensated transfer entropy of the EEG→EMG transmission direction. And the transfer entropy of the input direction EMG→EEG with delay compensation after delay compensation. ; (9)； (10)； In the formula, For delay The subsequent EMG embedding vector, For delay The EEG embedding vector is then used; ε represents the time interval. Indicates EMG signals at future moments The instantaneous amplitude or state; Indicates brainwave (EEG) signals at future moments The instantaneous amplitude or state; where: This is the current moment in the analysis, i.e., the reference point; This composite time point represents the exact future moment when the electromyographic or electroencephalographic result signal of interest should appear after compensating for the estimated, pathway-specific physiological conduction time. S33, Calculate the average delay-compensated transfer entropy, including: verifying the statistical significance of the delay-compensated transfer entropy in the same hemisphere and the same task phase based on the alternative data method to obtain all significant delay-compensated transfer entropies; averaging all significant delay-compensated transfer entropies to obtain the average delay-compensated transfer entropy. The average delay compensation transfer entropy As a quantitative index of CMC coupling strength, it is used to characterize the cortical-muscle functional coupling strength. The calculation formula is shown in Equation (11): (11)； In the formula, These represent the selected EEG channel set, EMG channel set, and sliding window set, respectively, where N(·) is the cardinality of the sets. This represents the optimal delay for the corresponding direction; S34, Construct a visualized, directionally delayed, compensated, directionally cortical-muscle connectivity (CMC) network, including: using EEG and EMG channels as nodes, and using the average delay to compensate for the transfer entropy. Using edge weights, a directional cortical-muscle connectivity (CMC) network with directional delay compensation is constructed, wherein the EEG channels represent the sensor's motor cortical region; S4. Extract assessment features and perform brain-muscle function change coupling analysis based on the assessment features; wherein, the assessment features are based on conduction delay, CMC coupling strength in hemispheric and task-specific stages, and hemispheric lateralization index. The data was extracted from four dimensions: the correlation between delayed-motion function behavioral index scores and other data.

2. The method for coupled analysis of brain-muscle function changes based on bidirectional delay compensation according to claim 1, characterized in that, S1 includes: S11, Collect EEG signals, EMG signals, and behavioral data from the test subjects; S12, preprocess the EEG signals, EMG signals, and behavioral data to remove noise and artifacts and retain effective neural electrical signals.

3. The method for coupled analysis of brain-muscle function changes based on bidirectional delay compensation according to claim 2, characterized in that, The preprocessing includes: (1) Perform signal downsampling, including: downsampling the EEG signal from 2000Hz to 1000Hz, downsampling the EMG signal from 2000Hz to 1000Hz, and downsampling the grip strength signal in the behavioral data from 200Hz to 1000Hz; (2) Noise filtering process, including: extracting the beta band signal using a 13-30Hz bandpass filter; removing power frequency noise using a 50Hz notch filter; wherein the beta band signal is the dominant frequency band of cortical muscle functional connectivity; (3) Removal of artifacts, including: based on independent component analysis combined with the ICLABEL plugin, identification and removal of electrooculography artifacts, electrocardiogram artifacts and electromyography interference artifacts in the EEG signal; (4) Perform signal rereference and segmentation: Rereference the EEG signal as the average reference of all channels; divide the grip force tracking period into resting phase S0, ramp-up phase S1 and maintenance phase S2 according to the contour of the grip force signal, and synchronously segment the EEG signal and the EMG signal according to the segmentation of the grip force tracking period. (5) Baseline correction: Using the phase signal of resting phase S0 as the baseline, the phase signals of ramp-up phase S1 and maintenance phase S2 are corrected.

4. The method for coupled analysis of brain-muscle function changes based on bidirectional delay compensation according to claim 3, characterized in that, S2 includes: S21, Set a sliding window, including: using an overlapping sliding window to divide the EEG signal and EMG signal after the synchronous segmentation; S22, perform phase space reconstruction, including: perform phase space reconstruction on the EEG signal and EMG signal in each sliding window to obtain equation (1) and equation (2): （1）； （2）； In the formula, This represents the time series of EEG signals. This represents the time series of electromyographic (EMG) signals, where τ represents the associated time delay, determined by the correlation integral method. , All represent the embedding dimension, determined by the Cao criterion; S23, Calculate the transfer entropy TE, including: calculating the transfer entropy in the outgoing direction EEG→EMG respectively. The transfer entropy with the input direction EMG→EEG ,in Characterizing the predictive ability of EEG signals to EMG signals. Transition entropy characterizes the predictive ability of electromyography (EMG) signals on electroencephalography (EEG) signals. and transfer entropy The calculation formulas are shown in (3) and (4) respectively: （3）； （4）； In the formula, H(·) is the conditional entropy, and ε represents the time interval. They are respectively Embedding vectors of EEG and EMG signals at any given time; and Both are scalar random variables, representing two time series, EEG signals and EMG signals, respectively, at a specific future moment. The observed values ​​or state, Indicates from the current reference time Beginning, moving towards the future A moment after that time; S24, Calculate the transfer entropy (DTE) with delay compensation, including: candidate delay based on the output direction. Time-align the signals to calculate the delay-compensated transfer entropy from EEG to EMG in the outgoing direction. And candidate delays based on the incoming direction Time-align the signals to calculate the delay-compensated transfer entropy from EMG to EEG in the incoming direction. The transfer entropy of the outgoing direction EEG→EMG with delay compensation and the transfer entropy of the input direction EMG→EEG with delay compensation The calculation formulas are shown in (5) and (6) respectively: (5)； (6)； In the formula, For sliding window index, and These are the channel indices for EEG and EMG signals, respectively. For delay Embedded vector of EMG signal after electromyography (EMG) For delay Embedded vectors of EEG signals after EEG; It is a scalar random variable, representing the value at the th . Within a sliding analysis window, data from electromyography channels... The signal at time point The instantaneous amplitude or state; It is a scalar random variable, representing the value at the th . Within a sliding analysis window, data from the EEG channel... The signal at the time point The instantaneous amplitude or state; S25, verifying the statistical significance of the delay-compensated transfer entropy DTE based on the substitute data method includes: generating the substitute data by performing phase randomization on the original EEG-EMG signal, and calculating the delay-compensated transfer entropy DTE value of the substitute data. The t-test was used to determine whether the delay-compensated transfer entropy (DTE) value of the original EEG-EMG signal was significantly higher than the delay-compensated DTE value of the surrogate data. The delay-compensated transfer entropy (DTE) value of the original EEG-EMG signal is significantly higher than that of the delay-compensated transfer entropy (DTE) value of the surrogate data. This indicates that there is a directed causal interaction between the channel pairs; S26, Perform bidirectional delay optimization, including: designing optimization criteria to maximize the sum of significant DTEs across EEG-EMG channel pairs and sliding windows, and searching for the optimal delay in the outgoing direction after optimization. And optimized input direction optimal delay The optimized outgoing direction optimal delay And optimized input direction optimal delay The optimization formulas are shown in equations (7) and (8): (7)； (8)； In the formula, and These are the channel indices for EEG and EMG signals, respectively. For sliding window index, This is the DTE change tracking function. The indicator function takes the value 1 when the delay-compensated transfer entropy (DTE) is significant, and 0 otherwise. , These are the estimated corticomuscular conduction times for the efferent pathway and the afferent pathway, respectively. It is a positive integer representing the summation symbol. The upper limit represents the total number of EEG signal channels in the system, channel index. Iterate from 1 to It covers all available EEG signal channels; It is a positive integer representing the summation symbol. The upper limit represents the total number of EMG signal channels in the system, channel index. Iterate from 1 to It covers all available electromyographic (EMG) signal channels; It is a positive integer representing the summation symbol. The upper limit represents the total number of sliding windows divided over the entire data time series; the sliding window index. Iterate from 1 to It covers all analysis time segments from the beginning to the end of the experiment.

5. The method for coupled analysis of brain-muscle function changes based on bidirectional delay compensation according to claim 4, characterized in that, Eighteen channels from the sensor's motion cortex region were selected as the EEG channels. These 18 channels include FC1, FC3, FC5, C1, C3, C5, CP1, CP3, and CP5 channels from the left hemisphere and FC2, FC4, FC6, C2, C4, C6, CP2, CP4, and CP6 channels from the right hemisphere. Based on the tester's target functional side, the 18 channels were categorized into target hemisphere channels and control hemisphere channels.

6. The method for coupled analysis of brain-muscle function changes based on bidirectional delay compensation according to claim 5, characterized in that, S4 includes: S41, extracting cortical muscle conduction time features based on the conduction delay dimension, including: extracting the optimized outward direction optimal delay. And optimized input direction optimal delay As a characteristic of cortical-muscular conduction time, it reflects the conduction efficiency of the neural pathway; S42, extract CMC coupling strength features based on the CMC coupling strength dimension of the hemisphere and task-specific phases, including: extracting the average delay compensation transfer entropy of the efferent and afferent directions of the target hemisphere and the control hemisphere in the resting phase S0, ramp-up phase S1 and maintenance phase S2 as the CMC coupling strength features, thereby reflecting the cortical muscle function interaction strength. S43, based on the hemispheric lateralization index Dimensional extraction of the hemispherical lateralization index Features include: calculating the hemispheric lateralization index based on the average delay-compensated transfer entropy of the target side and the control side. This quantifies the degree of lateralization in motion control functions, specifically the hemispherical lateralization index. The calculation formula is shown in equation (12): (12)； In the formula, The average delay compensation transfer entropy for the target hemisphere, For the control lateral hemisphere , The value range is [-1, 1], with positive values ​​indicating that the target hemisphere is dominant and negative values ​​indicating that the control hemisphere is dominant. S44, extract behavioral and motor function scoring features based on the correlation dimension of the delayed-motor function behavioral index score, including: taking the RMSE of the grip strength tracking task as the behavioral feature of fine motor control ability assessment, and taking the behavioral and motor function scores as the motor function scoring features. S45, Perform a coupling analysis of brain muscle function changes based on the assessment features.

7. A brain-muscle function change coupling analysis system based on bidirectional delay compensation, used to implement the method according to any one of claims 1-6, characterized in that, include: The data acquisition module (101) is used to design experimental paradigms and perform signal acquisition and signal preprocessing based on the experimental paradigms. The conduction time optimization module (102) is used to determine the bidirectional cortical muscle delay estimation algorithm based on the sliding window and the transfer entropy TE, and to quantify the conduction time of the EEG→EMG efferent pathway and the EMG→EEG infferent pathway of the cortical muscle of the test personnel based on the bidirectional cortical muscle delay estimation algorithm and the preprocessed signal. The network construction and coupling strength delay compensation module (103) is used to perform signal time alignment compensation based on the conduction time of the EEG→EMG outgoing path and the EMG→EEG incoming path, calculate the delay-compensated transfer entropy and the average delay-compensated transfer entropy, and construct a directional delay-compensated directional corticomuscular connection (CMC) network based on the delay-compensated transfer entropy and the average delay-compensated transfer entropy. The directional delay-compensated directional corticomuscular connection (CMC) network is used to visually demonstrate the strength distribution of bidirectional functional connections in the cortex and muscles. The network construction and coupling strength delay compensation module (103) is also used to: based on the optimized outgoing direction optimal delay... And optimized input direction optimal delay The EEG and EMG signals are time-aligned; the time-aligned EEG and EMG signals are then substituted into formulas (9) and (10) to calculate the delay-compensated transfer entropy of the EEG→EMG transmission direction. And the transfer entropy of the input direction EMG→EEG with delay compensation after delay compensation. ; (9)； (10)； In the formula, For delay The subsequent EMG embedding vector, For delay The EEG embedding vector is then used; ε represents the time interval. Indicates EMG signals at future moments The instantaneous amplitude or state; Indicates brainwave (EEG) signals at future moments The instantaneous amplitude or state; where: This is the current moment in the analysis, i.e., the reference point; and The optimal physiological conduction delay is estimated and obtained through a global search using the aforementioned optimization formulas (7) and (8). This composite time point represents the exact future moment when the expected EMG or EEG result signal should appear after compensating for the estimated, pathway-specific physiological conduction time. Calculating the average delay-compensated transfer entropy includes: verifying the statistical significance of the delay-compensated transfer entropy within the same hemisphere and the same task phase using surrogate data methods to obtain all significant delay-compensated transfer entropies; and averaging all significant delay-compensated transfer entropies to obtain the average delay-compensated transfer entropy. The average delay compensation transfer entropy As a quantitative index of CMC coupling strength, it is used to characterize the cortical-muscle functional coupling strength. The calculation formula is shown in Equation (11): (11)； In the formula, These represent the selected EEG channel set, EMG channel set, and sliding window set, respectively, where N(·) is the cardinality of the sets. To determine the optimal delay in the corresponding direction; a visualized, directional delay-compensated directional cortical-muscle connectivity (CMC) network is constructed, including: using the EEG and EMG channels as nodes, and compensating for the transfer entropy with the average delay. Using edge weights, a directional cortical-muscle connectivity (CMC) network with directional delay compensation is constructed, wherein the EEG channel is the sensor motion cortical region; The brain muscle function change coupling analysis module (104) is used to extract assessment features and perform brain muscle function change coupling analysis based on the assessment features; wherein, the assessment features are based on conduction delay, CMC coupling strength in hemispheric and task-specific stages, and hemispheric lateralization index. The data was extracted from four dimensions: the correlation between delayed-motion function behavioral index scores and other data.

8. An electronic device, characterized in that, It includes a processor and a memory, the memory storing multiple instructions, and the processor being used to read the instructions and execute the method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a plurality of instructions, which can be read by a processor and executed according to any one of claims 1-6.

Images