An electrical stimulation rehabilitation training system

Abstract

The application discloses a kind of based on electric stimulation rehabilitation training system, including virtual reality module, information acquisition module, information processing analysis module, task evaluation module, electric stimulation switch module, feedback module, electric stimulation control module and electric stimulation module.The application combines nerve information detection means and its response mechanism, strengthens the coupling effect of each feedback link, improves the matching efficiency of brain-muscle-limb multi-source information feedback and electric stimulation auxiliary rehabilitation training, so as to realize optimal motor rehabilitation feedback training effect.

Description

[0001] This application is a divisional application of the invention patent application with application number "202111486771.2", application date December 8, 2021, and invention title "An Electrical Stimulation Rehabilitation Training System Based on Multi-Source Information Coupling Feedback". Technical Field

[0002] This invention relates to the field of limb motor rehabilitation training, and in particular to an electrical stimulation rehabilitation training system based on multi-source information coupling feedback. Background Technology

[0003] Functional electrical stimulation (FES) is a type of neuromuscular electrical stimulation physical technology. It uses pre-designed low-frequency pulsed currents with specific waveforms, intensities, and repetition frequencies to stimulate specific muscle groups according to a predetermined program. This induces muscles to mimic normal voluntary movement or perform specific actions according to a treatment plan, accelerating the neuroplasticity process in stroke patients and gradually restoring limb motor function. Currently, FES in clinical rehabilitation is mostly passive, using fixed frequencies and durations for stimulation. It lacks feedback and real-time adjustment of multi-source information from the brain, muscles, and body, making it difficult to fully realize the maximum behavioral gain of FES. Post-stroke motor dysfunction is caused by neuromuscular pathway damage and abnormal muscle coordination due to cerebrovascular lesions. It is often accompanied by multi-level changes in information interaction characteristics, including oscillations between different brain regions, information transmission between the brain and limb muscles, synergistic effects between limb muscles, and functional coupling between neurovascular systems. The rehabilitation process requires the joint participation and coordination of multiple levels of units, including the brain, muscles, limbs, and physiological information. The interaction relationships under different movements also exhibit certain time-varying, bidirectional, and nonlinear coupling characteristics. Although current research on multimodal motor neurofeedback training has effectively overcome the above limitations to some extent, the synchronization and synergy between different feedback modes still need to be strengthened.

[0004] To address this, this invention proposes an electrostimulation rehabilitation training system based on multi-source information coupling feedback. By combining neural information detection methods and their response mechanisms, the coupling effect of each feedback link is strengthened, improving the matching efficiency between brain-muscle-limb multi-source information feedback and electrostimulation-assisted rehabilitation training, thereby achieving optimal motor rehabilitation feedback training results. Summary of the Invention

[0005] To achieve the objectives of this invention, the following technical solution is adopted:

[0006] An electrostimulation rehabilitation training system based on multi-source information coupling feedback includes a virtual reality module, an information acquisition module, an information processing and analysis module, a task evaluation module, an electrostimulation switch module, a feedback module, an electrostimulation control module, and an electrostimulation module. Specifically: the virtual reality module provides rehabilitation training tasks to the patient based on clinical assessment results; the information acquisition module acquires brain function, electrophysiological, and motion image information during the patient's rehabilitation training process; the information processing and analysis module processes and analyzes the brain function, electrophysiological, and motion image information synchronously acquired from the acquisition module; the task evaluation module evaluates the patient's task completion under virtual reality based on the information obtained from the information processing and analysis module; the electrostimulation switch module activates or deactivates electrostimulation based on the patient's task completion evaluation; the feedback module provides feedback on the coupling between brain function and electrocardiogram (ECG) and electromyography (EMG) during electrostimulation; the electrostimulation control module adjusts the electrostimulation control parameters based on the coupling information fed back from the feedback module; and the electrostimulation module activates or deactivates based on activation or deactivation commands output by the electrostimulation switch module and adjusts parameters based on specific parameter commands output by the electrostimulation control module.

[0007] The aforementioned electrostimulation rehabilitation training system based on multi-source information coupling feedback includes an information acquisition module comprising: a near-infrared brain function device for acquiring near-infrared photoneural signals of brain oxygenation during rehabilitation training; an electromyography (EMG) information acquisition device for acquiring EMG signals of the patient during rehabilitation training; an electrocardiogram (ECG) acquisition device for acquiring ECG signals of the patient during rehabilitation training; and a depth camera for acquiring limb movement image information of the patient during rehabilitation training.

[0008] The aforementioned electrostimulation rehabilitation training system based on multi-source information coupling feedback includes an information processing and analysis module for preprocessing the signals acquired by the information acquisition module. Specifically, the information processing and analysis module preprocesses the signals acquired by the information acquisition module by filtering the near-infrared light nerve signals of the brain to remove long-distance baseline drift and interference noise; processing the electromyographic signals using a Gaussian filter to remove power frequency noise; processing the electrocardiogram signals using a bandpass filter to remove artifacts; and preprocessing the limb movement image information.

[0009] The aforementioned electrostimulation rehabilitation training system based on multi-source information coupling feedback includes a task assessment module that establishes a patient task completion assessment model based on the patient's brain functional connectivity, focus, and task score during rehabilitation training.

[0010] The aforementioned electrostimulation rehabilitation training system based on multi-source information coupling feedback includes a task assessment module that establishes a patient task completion assessment model as follows:

[0011] Complex wavelet transform and wavelet phase coherence calculation were performed on the near-infrared photoneural signals of the brain of patients after pretreatment during rehabilitation training. The Pearson correlation coefficient and significance level between photoneural signals of each channel were calculated. If the photoneural signals of two channels were significantly correlated, functional connectivity was defined as existing. The number of functional connectivity channels in the healthy and affected brain regions was then calculated. Based on the calculation rules for the affected brain functional connectivity index, the affected brain functional connectivity index L was derived. i :

[0012]

[0013] Where CI represents the number of functional connections in the healthy brain region, TI represents the total number of pathways in the healthy brain region, CC represents the number of pathways for functional connections in the affected brain region, TC represents the total number of pathways in the affected brain region, and L represents the number of pathways. jk L represents the shortest path between any two channels, used to characterize the efficiency and rate of information transmission. jk In this context, j and k represent two different brain region pathways;

[0014] Heart rate variability (HRV) indices were extracted from the electrocardiogram (ECG) signals of pretreated patients during rehabilitation training, and time-domain and frequency-domain analyses were performed to establish an ECG-based attention index, Z0. a :

[0015]

[0016] Wherein, LF and HF are the low-frequency power and high-frequency power of the heart rate variability index, respectively, and P i (e jw δ(e) represents the average power spectrum of the heart rate variability signal over a certain sampling period. jw ) represents the power spectrum of the pulse function, SDNN is the standard deviation of the time-domain features of heart rate variability, PNN is the percentage of heart rate variability signals with an RR interval greater than 50 milliseconds out of the total number of RR signals, α and β are weighting coefficients, e is the natural constant, w is the center frequency of the heart rate variability signal, and j is the imaginary part of the complex number.

[0017] The motion image and electromyography signals of pretreated patients during rehabilitation training were compared and analyzed with the motion trajectories of the actual virtual reality task. A task scoring index P in the virtual reality environment was established. c :

[0018]

[0019] Where Q is the motion trajectory integrity coefficient, Rc is the system's comparison score with the virtual reality task trajectory, and X... i The signal is the preprocessed electromyography (EMG) signal, where N is the length of the time window corresponding to the EMG signal, n is the number of sampling points for the EMG signal, and θ, This represents the maximum joint movement angle of the shoulder and elbow joints in the depth camera image signal.

[0020] The aforementioned electrostimulation rehabilitation training system based on multi-source information coupling feedback includes a task assessment module that, based on a data-driven weighted analysis mechanism, establishes a patient rehabilitation training task completion index F from three aspects: brain functional connectivity indicators, attention indicators, and task scoring indicators.

[0021] F = C1 * L i +C2*Z a +C3*P c

[0022] Among them, C1, C2, and C3 are weight coefficients.

[0023] The data-driven weighting analysis mechanism is as follows:

[0024]

[0025] Among them, C j p represents the weight coefficient of the j-th feature. j The standard deviation of the j-th feature represents the degree of fluctuation of the j-th feature within the collected sample database of patient rehabilitation training task completion. Let x be the n sample values ​​of the j-th feature. j (1) x j (2)...x j (n),

[0026]

[0027] w j The deviation coefficient can be obtained from the following formula:

[0028]

[0029] Among them, w j This represents the deviation coefficient, and λ is the threshold value.

[0030] The aforementioned electrostimulation rehabilitation training system based on multi-source information coupling feedback, wherein the feedback module represents the coupling strength of different signals by calculating the coherence of different signals according to the power spectrum calculation method:

[0031] Brain-brain coupling strength:

[0032]

[0033] Among them, C nn The coupling strength of different brain regions is represented by , m is the number of functionally connected channels on the affected side after electrical stimulation, and CP is... ij(w) represents the power spectrum of brain oxygenation signals in different channels, CP ii (w) represents the autopower spectrum of the brain oxygenation signal in channel i, CP jj (w) represents the autopower spectrum of the brain oxygen signal in channel j;

[0034] Heart-brain coupling strength:

[0035]

[0036] Among them, C nx The coupling strength between cerebral oxygenation signals and heart rate variability in different channels is represented by m, where m is the number of functionally connected channels on the affected side after electrical stimulation, and CP is the number of channels. ix (w) represents the power spectra of brain oxygenation signals and heart rate variability time series from different channels, CP ii (w) represents the autopower spectrum of the brain oxygenation signal in channel i, CP xx (w) represents the autopower spectrum of the heart rate variability time series;

[0037] Brain-muscle coupling strength:

[0038]

[0039] Among them, C nz The coupling strength between cerebral oxygenation signals and electromyography in different channels is represented by m, where m is the number of functionally connected channels on the affected side after electrical stimulation, and CP is the coefficient of performance. iz (w) represents the power spectra of brain oxygenation and electromyography signals in different channels over time. ii (w) represents the autopower spectrum of the brain oxygenation signal in channel i, CP zz (w) represents the autopower spectrum of the electromyographic signal.

[0040] The aforementioned electrostimulation rehabilitation training system based on multi-source information coupling feedback includes the following steps: When a patient uses the electrostimulation module for the first time, the electrostimulation control module pre-sets the frequency, pulse width, and amplitude of the electrostimulation module, and determines whether the brain-brain coupling strength index exceeds threshold M1. If it does not exceed the threshold, the frequency of the electrostimulation module is increased; if it exceeds the normal threshold, the frequency of the electrostimulation module is decreased. Next, it determines whether the heart-brain coupling strength index exceeds threshold M2. If it does not exceed the threshold, the pulse width of the electrostimulation module is increased; if it exceeds the normal threshold, the pulse width of the electrostimulation module is decreased. Finally, it determines whether the brain-muscle coupling strength index exceeds threshold M3. If it does not exceed the threshold, the amplitude of the electrostimulation module is increased; if it exceeds the normal threshold, the amplitude of the electrostimulation module is decreased.

[0041] The aforementioned electrostimulation rehabilitation training system based on multi-source information coupling feedback includes: an electrostimulation control module that continuously collects data such as patient basic information, brain-brain coupling strength indicators, heart-brain coupling strength indicators, and brain-muscle coupling strength, along with corresponding optimal electrostimulation parameters, to establish an electrostimulation parameter adjustment database. It then utilizes artificial intelligence algorithms to intelligently output specific values ​​for the electrostimulation parameters, improving the efficiency of rehabilitation training. Based on the electrostimulation parameter adjustment database, the electrostimulation control module establishes a long short-term memory neural network model.

[0042] [W,M,F,T]=G LSTM (B,P,C nn C nx C nz )

[0043] Where W, M, F, and T are the frequency, pulse width, amplitude, and time parameters output by the electrical stimulation module, and G... LSTM To train a long short-term memory neural network model, B represents the patient's basic information, and P and C... nn C nx C nz Thresholds for clinical assessment results, brain-brain coupling strength, heart-brain coupling strength, and brain-muscle coupling strength indicators at different stages of patient rehabilitation. Attached Figure Description

[0044] Figure 1 This is a schematic diagram of the electrical stimulation rehabilitation training system based on multi-source information coupling feedback according to the present invention.

[0045] Figure 2 This is a flowchart of the electrostimulation rehabilitation training method based on multi-source information coupling feedback according to the present invention;

[0046] Figure 3 This is a schematic diagram illustrating the adjustment of electrical stimulation parameters according to the present invention. Detailed Implementation

[0047] The following is in conjunction with the appendix Figure 1-3 The specific embodiments of the present invention will be described in detail below.

[0048] In the following description, only certain exemplary embodiments are briefly described. As those skilled in the art will recognize, the described embodiments can be modified in various ways without departing from the spirit or scope of the invention. Therefore, the drawings and description are considered to be exemplary in nature and not restrictive.

[0049] like Figure 1As shown, the present invention discloses an electrostimulation rehabilitation training system based on multi-source information coupling feedback, comprising: a virtual reality module, an information acquisition module, an information processing and analysis module, a task assessment module, an electrostimulation switch module, a feedback module, an electrostimulation control module, and an electrostimulation module. Wherein:

[0050] The virtual reality module provides rehabilitation training tasks for patients based on their clinical assessment results; the information acquisition module collects brain function, electrophysiological, and motion image information during the rehabilitation training process; the information processing and analysis module processes and analyzes the brain function, electrophysiological, and motion image information synchronously acquired from the acquisition module; the task assessment module evaluates the patient's task completion under virtual reality based on the information obtained from the information processing and analysis module; the electrical stimulation switch module activates or deactivates the electrical stimulation module based on the patient's task completion assessment; the feedback module provides feedback on the coupling between brain function and electrocardiogram (ECG) and electromyography (EMG) during the electrical stimulation process; the electrical stimulation control module outputs electrical stimulation control parameter commands based on the coupling information fed back from the feedback module; and the electrical stimulation module activates or deactivates based on the activation or deactivation commands output by the electrical stimulation switch module and adjusts parameters based on the specific parameter commands output by the electrical stimulation control module.

[0051] The virtual reality module is used to provide rehabilitation training tasks to patients based on their clinical scale (motor scale) assessment results. Preferably, the difficulty level of the rehabilitation training tasks should be adjusted according to the patient's assessment, and the patient should be able to complete at least 50% of the rehabilitation training tasks at different stages.

[0052] The information acquisition module includes: a near-infrared brain function device for acquiring near-infrared photoneural signals of brain oxygenation during rehabilitation training; an electromyography (EMG) information acquisition device for acquiring EMG signals of the patient during rehabilitation training; an electrocardiogram (ECG) acquisition device for acquiring ECG signals of the patient during rehabilitation training; and a depth camera for acquiring limb movement image information of the patient during rehabilitation training.

[0053] The information processing and analysis module is used to preprocess the signals acquired by the information acquisition module, namely: filtering the near-infrared light nerve signals of brain oxygenation to remove long-distance baseline drift and interference noise; processing the electromyography signals with a Gaussian filter to remove power frequency noise; processing the electrocardiogram signals with a bandpass filter to remove artifacts; and processing the limb motion image information with median filtering to remove isolated noise.

[0054] The task assessment module establishes a model for assessing patient task completion based on brain functional connectivity, focus, and task scores during rehabilitation training.

[0055] The task assessment module establishes a patient task completion assessment model as follows:

[0056] Step 1. Perform complex wavelet transform and wavelet phase coherence calculation on the near-infrared photoneural signals of the brain during rehabilitation training of the pre-treated patients. Calculate the Pearson correlation coefficient and significance level between the photoneural signals of each channel. If the photoneural signals of two channels are significantly correlated, functional connectivity is defined as existing. Then, calculate the number of functional connectivity channels in the healthy and affected brain regions, and based on the calculation rules for the affected brain functional connectivity index, derive the affected brain functional connectivity index L. i :

[0057]

[0058] Where CI represents the number of functional connections in the healthy brain region, TI represents the total number of pathways in the healthy brain region, CC represents the number of pathways for functional connections in the affected brain region, TC represents the total number of pathways in the affected brain region, and L represents the number of pathways. jk L represents the shortest path between any two channels, used to characterize the efficiency and rate of information transmission. jk In this context, j and k represent two distinct brain pathways. L i The value ranges from 0 to 1, where 0 indicates that the affected side has no brain functional network connection, and 1 indicates that the brain functional connections on the affected side and the healthy side are symmetrical.

[0059] Step 2. Extract heart rate variability (HRV) from the electrocardiogram (ECG) signals of pretreated patients during rehabilitation training, and perform time-domain and frequency-domain analyses to establish an ECG-based attention index, Z0. a :

[0060]

[0061] Wherein, LF and HF are the low-frequency power and high-frequency power of the heart rate variability index, respectively, and P i (e jw δ(e) represents the average power spectrum of the heart rate variability signal over a certain sampling period. jw Z represents the power spectrum of the pulse function, SDNN is the standard deviation of the time-domain characteristics of heart rate variability, PNN is the percentage of heart rate variability signals with an RR interval greater than 50 milliseconds out of the total number of RR signals, α and β are weighting coefficients, e is the natural constant, w is the center frequency of the heart rate variability signal, and j is the imaginary part of the complex number. a The value ranges from 0 to 1, where (0-0.5) indicates inattentiveness and insufficient focus, [0.5-0.8) indicates focused, relaxed, and highly focused attention, and [0.8-1] indicates highly focused, tense, and extremely focused attention.

[0062] Step 3. Analyze the motion image and electromyographic signals of the pre-processed patient during rehabilitation training, comparing the patient's motion trajectory with the actual motion trajectory of the virtual reality task, and establish a task scoring index P in the virtual reality environment. c :

[0063]

[0064] Where Q is the motion trajectory integrity coefficient, Rc is the system's comparison score with the virtual reality task trajectory, and X... i The signal is the preprocessed electromyography (EMG) signal, where N is the time window length for EMG signal sampling, n is the number of EMG signal sampling points, and θ, P represents the maximum joint range of motion of the shoulder and elbow joints obtained from the depth camera image signal. c The value ranges from 0 to 1, P c The smaller the value, the lower the score for task completion; P c The higher the value, the higher the score for task completion.

[0065] Step 4. Based on a data-driven weighting analysis mechanism, establish a patient rehabilitation training task completion index F from three aspects: brain functional connectivity index, attention index, and task scoring index.

[0066] F = C1 * L i +C2*Z a +C3*P c

[0067] Among them, C1, C2, and C3 are weight coefficients.

[0068] The data-driven weighting analysis mechanism is as follows:

[0069]

[0070] Among them, C j p represents the weight coefficient of the j-th feature. j This represents the degree of fluctuation of the j-th feature within the collected sample database of patient rehabilitation training task completion, i.e., the standard deviation of the j-th feature. For example, x is the n sample values ​​of the j-th feature. j (1) x j (2)...x j (n),

[0071]

[0072] w j The deviation coefficient can be obtained from the following formula:

[0073]

[0074] Among them, w j This represents the deviation coefficient, and λ is the threshold value.

[0075] The electrical stimulation switch module is used to determine whether the electrical stimulation module needs to be activated based on the patient's task completion assessment during rehabilitation training.

[0076] When the patient's rehabilitation training task completion rate (F) falls below 50%, the electrical stimulation switch module activates the electrical stimulation module, which then provides electrical stimulation therapy to the patient according to preset electrical stimulation parameters and duration. For example, the initial activation time of the electrical stimulation module during the entire rehabilitation training is 5 minutes, and subsequent activations will be adjusted based on specific circumstances.

[0077] The feedback module receives the pre-processed signals from the information processing and analysis module, calculates the brain-brain coupling strength, heart-brain coupling strength, and brain-muscle coupling strength indices during the patient's electrical stimulation rehabilitation training, and outputs them to the electrical stimulation control module.

[0078] The brain-brain coupling strength index is used to assess the functional connectivity of the affected brain region during electrical stimulation; the heart-brain coupling strength index is used to assess the brain's control over focus during electrical stimulation; and the brain-muscle coupling index is used to assess the coordinated regulation of muscles and the brain during electrical stimulation.

[0079] The feedback module uses a power spectrum calculation method to represent the coupling strength of different signals by calculating the coherence of different signals, as detailed below:

[0080] Brain-brain coupling strength:

[0081]

[0082] Among them, C nn The coupling strength of different brain regions is represented by , m is the number of functionally connected channels on the affected side after electrical stimulation, and CP is... ij (w) represents the power spectrum of brain oxygenation signals in different channels, CP ii (w) represents the autopower spectrum of the brain oxygenation signal in channel i, CP jj (w) represents the autopower spectrum of the brain oxygen signal in channel j.

[0083] Heart-brain coupling strength:

[0084]

[0085] Among them, C nx The coupling strength between cerebral oxygenation signals and heart rate variability in different channels is represented by m, where m is the number of functionally connected channels on the affected side after electrical stimulation, and CP is the number of channels. ix (w) represents the power spectra of brain oxygenation signals and heart rate variability time series from different channels, CP ii(w) represents the autopower spectrum of the brain oxygenation signal in channel i, CP xx (w) represents the autopower spectrum of the heart rate variability time series.

[0086] Brain-muscle coupling strength:

[0087]

[0088] Among them, C nz The coupling strength between cerebral oxygenation signals and electromyography in different channels is represented by m, where m is the number of functionally connected channels on the affected side after electrical stimulation, and CP is the coefficient of performance. iz (w) represents the power spectra of brain oxygenation and electromyography signals in different channels over time. ii (w) represents the autopower spectrum of the brain oxygenation signal in channel i, CP zz (w) represents the autopower spectrum of the electromyographic signal.

[0089] The electrical stimulation control module is used to output control commands for key parameters such as frequency, pulse width, and amplitude of the electrical stimulation module based on the brain-brain coupling strength, heart-brain coupling strength, and brain-muscle coupling strength indicators fed back by the feedback module.

[0090] The electrical stimulation module adjusts specific parameters based on the key parameter control commands output by the electrical stimulation control module, such as the frequency, pulse width, and amplitude of the electrical stimulation.

[0091] Studies have shown that when the pulse width and amplitude of electrical stimulation remain constant, electrical stimulation at different frequencies has a positive correlation with the connectivity of brain function in stroke patients; when the frequency and amplitude of electrical stimulation remain constant, electrical stimulation with different pulse widths has a positive correlation with the brain's control over attention; and when the frequency and pulse width of electrical stimulation remain constant, electrical stimulation with different amplitudes has a positive correlation with the brain's coordinated control over muscles. In other words, within a certain threshold range of brain-brain coupling strength, increasing the frequency of electrical stimulation enhances the connectivity of brain function; within a certain threshold range of heart-brain coupling strength, increasing the pulse width of electrical stimulation enhances the patient's attention; and within a certain threshold range of brain-muscle coupling strength, increasing the amplitude of electrical stimulation enhances the patient's brain-muscle coordinated control.

[0092] Preferably, the electrical stimulation module adaptively adjusts the frequency, pulse width, and amplitude of the output to the electrical stimulation module based on the brain-brain coupling strength, heart-brain coupling strength, and brain-muscle coupling strength indicators fed back by the feedback module. This allows the electrical stimulation-assisted rehabilitation training to achieve the maximum gain effect and improve the efficiency and effectiveness of the patient's electrical stimulation rehabilitation training.

[0093] like Figure 3As shown, the specific adjustments are as follows: When a patient uses the electrical stimulation module for the first time, the electrical stimulation control module pre-sets the frequency, pulse width, and amplitude of the electrical stimulation module, and determines whether the brain-brain coupling strength index exceeds the threshold M1. If it does not exceed the threshold, the frequency of the electrical stimulation module is increased; if it exceeds the normal threshold, the frequency of the electrical stimulation module is decreased. Then, it determines whether the heart-brain coupling strength index exceeds the threshold M2. If it does not exceed the threshold, the pulse width of the electrical stimulation module is increased; if it exceeds the normal threshold, the pulse width of the electrical stimulation module is decreased. Then, it determines whether the brain-muscle coupling strength index exceeds the threshold M3. If it does not exceed the threshold, the amplitude of the electrical stimulation module is increased; if it exceeds the normal threshold, the amplitude of the electrical stimulation module is decreased.

[0094] In the rehabilitation training task, the initial electrical stimulation time is set to 5 minutes, followed by a 1-minute rest before resuming the rehabilitation training. If the patient's task completion rate reaches more than 60% when the rehabilitation training is resumed, it indicates that the electrical stimulation effect is obvious. If the patient's task completion rate is still below 50%, the duration of the next electrical stimulation will be increased, for example, by 2 minutes each time.

[0095] After the electrical stimulation module is executed during rehabilitation training, the parameters such as the electrical stimulation frequency, pulse width, amplitude, and time at which the patient achieves the highest task completion rate are recorded and saved as the initial values ​​for the electrical stimulation parameters in the next rehabilitation training session.

[0096] The thresholds M1, M2, and M3 of the brain-brain coupling strength index, heart-brain coupling strength index, and brain-muscle coupling strength index are mainly determined by the maximum frequency, pulse width, and amplitude that the patient can tolerate when receiving resting-state electrical stimulation before each rehabilitation training session.

[0097] In addition, the electrical stimulation control module continuously collects data indicators such as the patient's basic information, brain-brain coupling strength index, heart-brain coupling strength index, and brain-muscle coupling strength and corresponding optimal electrical stimulation parameters to establish an electrical stimulation parameter adjustment database. It also uses artificial intelligence algorithms to intelligently output the specific values ​​of electrical stimulation parameters, thereby improving the efficiency of rehabilitation training.

[0098] Preferably, a long short-term memory neural network model is established based on the electrical stimulation parameter adjustment database. The model is continuously trained and tested to optimize the parameters of the neural network model, forming an intelligent recommendation model for electrical stimulation parameters. This model can recommend personalized adaptive electrical stimulation parameters and make fine adjustments based on the patient's real-time rehabilitation training, reducing the number of times parameters need to be manually adjusted.

[0099] [W,M,F,T]=G LSTM (B,P,C nn C nx C nz )

[0100] Where W, M, F, and T are the frequency, pulse width, amplitude, and time parameters output by the electrical stimulation module, and G... LSTM To train the long short-term memory neural network model effectively, B represents the patient's basic information, such as hemiplegia status and age, while P and C represent the patient's basic information. nn C nx C nz Thresholds for clinical assessment results, brain-brain coupling strength, heart-brain coupling strength, and brain-muscle coupling strength indicators at different stages of patient rehabilitation.

[0101] The rehabilitation training system also includes display devices, such as displays, for presenting the feedback process of the feedback module in real time.

[0102] Compared with the prior art, the beneficial effects of the present invention are:

[0103] (1) By establishing a patient rehabilitation training task completion index through three aspects: brain function connectivity index, attention index and task scoring index, the patient's rehabilitation training task completion index is fully considered. The collaborative control of multiple sources of information such as brain, physiological information and muscle during the patient's rehabilitation training process can be used to evaluate the patient's limb function in real time.

[0104] (2) Make full use of the multi-level information interaction characteristics of the patient’s brain-brain coupling, heart-brain coupling, brain-muscle coupling and other coupling characteristic indicators to adjust the electrical stimulation parameters in the patient’s rehabilitation training in real time, and promote the coordinated optimization and real-time feedback of brain-limb and electrical stimulation data.

[0105] (3) Using this system, patients can be provided with personalized and adaptive electrical stimulation parameter adjustment schemes, so that electrical stimulation-assisted rehabilitation training can achieve the maximum benefit effect and improve the efficiency and effectiveness of patients' electrical stimulation rehabilitation training.

[0106] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the scope of the technology disclosed in the present invention, or make equivalent substitutions for some of the technical features. Such modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. An electrical stimulation rehabilitation training system, comprising a virtual reality module, an information acquisition module, an information processing and analysis module, a task assessment module, an electrical stimulation switch module, a feedback module, an electrical stimulation control module, and an electrical stimulation module, characterized in that: The virtual reality module provides rehabilitation training tasks for patients based on their clinical assessment results; the information acquisition module collects brain function, electrophysiological, and motion image information during the rehabilitation training process; the information processing and analysis module processes and analyzes the brain function, electrophysiological, and motion image information synchronously acquired from the acquisition module; the task assessment module evaluates the patient's task completion under virtual reality based on the information obtained from the information processing and analysis module; the electrical stimulation switch module activates or deactivates electrical stimulation based on the patient's task completion assessment; the feedback module provides feedback on the coupling between brain function and electrocardiogram (ECG) and electromyography (EMG) during electrical stimulation; the electrical stimulation control module adjusts the electrical stimulation control parameters based on the coupling information fed back from the feedback module; and the electrical stimulation module activates or deactivates based on the activation or deactivation commands output by the electrical stimulation switch module and adjusts parameters based on the specific parameter commands output by the electrical stimulation control module. Specifically, the task assessment module, based on a data-driven weighted analysis mechanism, establishes a patient rehabilitation training task completion index F from three aspects: brain function connectivity indicators, attention indicators, and task scoring indicators. F=C1*L i +C2*Z a +C3*P c Among them, C1, C 2、 C3 is the weighting coefficient; The feedback module uses a power spectrum calculation method to represent the coupling strength of different signals by calculating the coherence of different signals: Brain-brain coupling strength: Among them, C nn The coupling strength of different brain regions is represented by , m is the number of functionally connected channels on the affected side after electrical stimulation, and CP is... ij (w) represents the power spectrum of brain oxygenation signals from different channels, CP ii (w) represents the autopower spectrum of the brain oxygenation signal in channel i, CP jj (w) represents the autopower spectrum of the brain oxygen signal in channel j; Heart-brain coupling strength: Among them, C nx The coupling strength between cerebral oxygenation signals and heart rate variability in different channels is represented by m, where m is the number of functionally connected channels on the affected side after electrical stimulation, and CP is the number of channels. ix (w) represents the power spectra of brain oxygenation signals and heart rate variability time series from different channels, CP ii (w) represents the autopower spectrum of the brain oxygenation signal in channel i, CP xx (w) represents the autopower spectrum of the heart rate variability time series; Brain-muscle coupling strength: Among them, C nz The coupling strength between cerebral oxygenation signals and electromyography in different channels is represented by m, where m is the number of functionally connected channels on the affected side after electrical stimulation, and CP is the coefficient of performance. iz (w) represents the power spectra of brain oxygenation and electromyography signals in different channels over time. ii (w) represents the autopower spectrum of the brain oxygenation signal in channel i, CP zz (w) represents the autopower spectrum of the electromyographic signal; When a patient uses the electrical stimulation module for the first time, the electrical stimulation control module pre-sets the frequency, pulse width, and amplitude of the electrical stimulation module, and judges the brain... If the brain coupling strength index does not exceed the threshold M1, the frequency of the electrical stimulation module is increased; if it exceeds the normal threshold, the frequency of the electrical stimulation module is decreased; then the heart... If the brain coupling strength index does not exceed the threshold M2, the pulse width of the electrical stimulation module is increased; if it exceeds the normal threshold, the pulse width of the electrical stimulation module is decreased; then the brain is reassessed. If the muscle coupling strength index does not exceed the threshold M3, the amplitude of the electrical stimulation module will be increased; if it exceeds the normal threshold, the amplitude of the electrical stimulation module will be decreased.

2. The electrical stimulation rehabilitation training system according to claim 1, characterized in that: The feedback module uses the power spectrum calculation method to calculate the coherence of different signals to represent the coupling strength of different signals.

3. The electrical stimulation rehabilitation training system according to claim 1, characterized in that: The electrical stimulation control module continuously collects basic patient information and brain... Brain coupling strength index, heart Brain coupling strength index and brain A database for adjusting electrical stimulation parameters is established using data indicators such as muscle coupling strength and corresponding optimal electrical stimulation parameters. Artificial intelligence algorithms are then used to intelligently output the specific values ​​of electrical stimulation parameters, thereby improving the efficiency of rehabilitation training. The electrical stimulation control module establishes a long short-term memory neural network model based on the database for adjusting electrical stimulation parameters.

Images