Rehabilitation system and method based on motor imagery electroencephalogram signals and eyeball tracking
By combining motor imagery EEG signals and eye-tracking technology, a system for autonomously selecting rehabilitation training modes was designed, which solves the problem of insufficient patient initiative in traditional rehabilitation training methods and improves rehabilitation effectiveness and classification accuracy.
Patent Information
- Application Number
- CN202511097254.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-06
- Publication Date
- 2025-12-16
AI Technical Summary
Existing rehabilitation training methods lack active patient participation, and traditional methods suffer from limited resources or high costs. Furthermore, non-invasive brain-computer interfaces have long training cycles and are tedious in rehabilitation training, affecting patient motivation and recognition accuracy.
Combining motor imagery EEG signals and eye-tracking technology, the system enables autonomous selection of rehabilitation training modes through signal acquisition, processing, eye tracking, and electrical stimulator modules. It collects signals using an eye tracker and EEG cap, classifies them using support vector machines and co-space patterns, and outputs stimulating pulse currents for rehabilitation training.
It increases patients' initiative and enthusiasm for rehabilitation training, improves the effect of rehabilitation treatment, and makes up for the shortcomings of limited classification patterns and low real-time classification accuracy of motor imagery EEG signals.
Smart Images
Figure CN121130288A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of rehabilitation technology, and specifically to a rehabilitation system and method based on motor imagery EEG signals and eye tracking. Background Technology
[0002] Limb paralysis caused by damage to the central or peripheral nervous system is commonly due to stroke, spinal cord injury, and other causes. Stroke, in particular, due to various reasons, severely endangers physical health, characterized by high morbidity, disability, recurrence, and mortality rates. Limb paralysis not only affects patients' daily lives and hinders their reintegration into society but also causes psychological trauma, impacting the effectiveness of rehabilitation treatment. Therefore, finding a rehabilitation training method that allows limb paralysis patients to participate independently in recovery exercises is a pressing issue that needs to be addressed.
[0003] Currently, there are two main categories of traditional rehabilitation training methods. One is rehabilitation training with the help of a therapist or family member, but this has the disadvantages of limited therapist resources and high requirements for rehabilitation facilities. The other is the assistance of mechanical exoskeleton devices, but most exoskeleton devices are expensive, increasing the cost of treatment. Moreover, both of these rehabilitation training methods lack the patient's active participation.
[0004] Functional electrical stimulation (FES) applies low-frequency pulsed currents of a certain intensity to the nerves surrounding specific muscles through a pre-programmed procedure, inducing muscle contraction to simulate normal voluntary movement. This helps hemiplegic patients enhance muscle function, improve joint range of motion, and thus achieve limb function reconstruction and improve the patient's quality of life.
[0005] Brain-computer interface (BCI) technology decodes a subject's thoughts and translates them into real-time command signals to control external devices. There are two main types: invasive and non-invasive. Invasive BCIs require surgical implantation of electrodes into the cerebral cortex, which can obtain high-quality neural signals and has significant therapeutic effects. However, the surgery carries high safety risks, potentially leading to complications such as infection and bleeding, and requires long-term maintenance after implantation. Non-invasive BCIs do not require implanted electrodes to collect signals; instead, they collect signals by wearing an EEG cap attached to the scalp. This avoids expensive and high-risk surgery and is simple and convenient to operate. Currently, there are studies combining BCIs with functional electrical stimulation (fESPS) for rehabilitation training. However, using BCIs in rehabilitation training still faces challenges such as long training cycles, monotonous processes, patient fatigue, and reduced BCI recognition accuracy, severely impacting patients' initiative and enthusiasm during rehabilitation.
[0006] Based on this, the present invention designs a rehabilitation system and method based on motor imagery EEG signals and eye tracking to solve the above problems. Summary of the Invention
[0007] In view of the above-mentioned shortcomings of the existing technology, the present invention provides a rehabilitation system and method based on motor imagery EEG signals and eye tracking.
[0008] To achieve the above objectives, the present invention provides the following technical solution:
[0009] Rehabilitation systems based on motor imagery EEG signals and eye tracking include:
[0010] Signal acquisition module: used to acquire EEG signals for motor imagery and record limb state information;
[0011] Signal processing module: used to store the collected motor imagery EEG signals, determine the motor intention through the motor imagery EEG signals, and output selection instructions;
[0012] Eye-tracking module: Used to track eye gaze and interface prompts, and coordinate with selection commands to issue treatment instructions;
[0013] Electrical stimulator module: used to receive treatment commands and output stimulation pulse current according to the treatment commands;
[0014] The host computer module is electrically connected to the signal acquisition module, eye tracking module, signal processing module, and electrical stimulator module. It is used to transmit the motor imagery EEG signals acquired by the signal acquisition module and the limb state information of the corresponding joints to the signal processing module, to transmit the selection instructions processed by the signal processing module to the eye tracking module, and to transmit the treatment instructions of the eye tracking module to the functional electrical stimulator module.
[0015] Display module: Used to display the selection interface and limb status information. The display module is connected to the host computer.
[0016] Furthermore, the signal acquisition module includes an EEG signal acquisition module for acquiring EEG signals for motor imagery and a sensor for acquiring limb state information. Both the EEG signal acquisition module and the angular acceleration sensor are connected to the host computer.
[0017] Furthermore, the specific operation of the signal processing module is as follows:
[0018] Step 1: A Butterworth filter of 8-30Hz is used to filter out noise from the motor imagery EEG signal to obtain the effective signal;
[0019] Step 2: Use Independent Component Analysis (ICA) or wavelet transform to remove artifacts such as electrooculography (EOG) and electromyography (EMG) from the effective signal to obtain the true signal;
[0020] Step 3: Train the spatial domain feature vector of the real signal using the Co-space Pattern (CSP) method;
[0021] Step 4: Classify the spatial domain feature vectors using Support Vector Machine (SVM), Linear Discriminant Analysis (LDA), or Multilayer Perceptron (MLP);
[0022] Step 5: The classification results are evaluated using ROC curves. The evaluation results are then fed into a real-time support vector machine (SVM) model to determine the motion intent. Based on the motion intent, selection instructions are output.
[0023] Furthermore, the selection command is either to move left or to remain still.
[0024] Furthermore, the eye-tracking module uses an eye tracker, which is fixed below the computer screen.
[0025] Furthermore, the eye-tracking module operates as follows:
[0026] Step 1: The eye tracker tracks the eye's gaze. The first selection interface displays the main category prompts. The first prompt is determined, and then it is determined whether the selection instruction is a leftward instruction. If it is, proceed to Step 2; if it is not, stop the operation.
[0027] Step 2: The display interface enters the second selection interface. The eye tracker tracks the eye's gaze and displays the detailed category prompts on the second selection interface to determine the second prompt instruction. Then, it determines whether the selection instruction is a leftward instruction. If the determination is yes, the second prompt instruction is converted into a treatment instruction and sent to the host computer. If the determination is no, the operation stops.
[0028] Furthermore, the electrical stimulator module includes a main control chip, a power supply module, and a constant current source module, with the main control chip electrically connected to the power supply module and the constant current source module.
[0029] An application method for a rehabilitation system based on motor imagery EEG signals and eye tracking is described below:
[0030] Step 1: Set the initial electrical stimulator current parameters and limb initial state information parameters of the electrical stimulator module;
[0031] Step 2: The subject sits in front of the display module and wears the EEG signal acquisition module. The EEG signal acquisition module collects EEG signals of motor imagery in the parietal lobe related to the sensorimotor cortex. The electrical stimulator module is attached to the treatment site.
[0032] Step 3: The signal processing module and the eye-tracking module transmit the treatment instructions to the host computer;
[0033] Step 4: The host computer of the angular acceleration sensor transmits the treatment command to the electrical stimulator module, the electrical stimulator module performs electrical stimulation treatment, and the angular acceleration sensor collects the subject's limb movement status information;
[0034] Step 5: Calculate the difference between the actual limb movement state information parameters and the limb initial state information parameters. If the difference is ≥0, the stimulation ends, and the initial electrical stimulator current parameters and limb initial state parameters are adjusted. If the difference is <0, treatment continues.
[0035] Beneficial effects: This invention combines real-time acquisition of non-invasive motor imagery signals with eye-tracking technology, allowing subjects to autonomously select rehabilitation training modes for different body parts based on their intentions. This increases patients' initiative and enthusiasm during rehabilitation training, and compensates for the shortcomings of limited classification modes and low real-time classification accuracy of motor imagery EEG signals, thus effectively improving the rehabilitation treatment effect for patients. Attached Figure Description
[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.
[0037] Figure 1 This is a schematic diagram of the category prompt instructions displayed on the first selection interface of the present invention;
[0038] Figure 2 The second prompt instruction displayed in the second selection interface of the present invention is a wrist joint diagram;
[0039] Figure 3 The second prompt instruction displayed in the second selection interface of the present invention is an elbow joint diagram;
[0040] Figure 4 The second prompt instruction for the detailed category prompt instruction displayed in the second selection interface of the present invention is a schematic diagram of the wrist + elbow joint. Detailed Implementation
[0041] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.
[0042] The present invention will be further described below with reference to embodiments.
[0043] Example 1: A rehabilitation system based on motor imagery EEG signals and eye tracking, comprising:
[0044] Signal acquisition module: used to acquire EEG signals for motor imagery and record limb state information;
[0045] Signal processing module: used to store the collected motor imagery EEG signals, determine the motor intention through the motor imagery EEG signals, and output selection instructions;
[0046] Eye-tracking module: Used to track eye gaze and interface prompts, and coordinate with selection commands to issue treatment instructions;
[0047] Electrical stimulator module: used to receive treatment commands and output stimulation pulse current according to the treatment commands;
[0048] The host computer module is electrically connected to the signal acquisition module, eye tracking module, signal processing module, and electrical stimulator module. It is used to transmit the motor imagery EEG signals acquired by the signal acquisition module and the limb state information of the corresponding joints to the signal processing module, to transmit the selection instructions processed by the signal processing module to the eye tracking module, and to transmit the treatment instructions of the eye tracking module to the functional electrical stimulator module.
[0049] Display module: Used to display the selection interface and limb status information. The display module is connected to the host computer.
[0050] This invention combines real-time acquisition of non-invasive motor imagery signals with eye-tracking technology, allowing subjects to autonomously select rehabilitation training modes for different body parts based on their intentions. This increases patients' initiative and enthusiasm during rehabilitation training, and compensates for the shortcomings of limited classification modes and low real-time classification accuracy in motor imagery EEG signals, thus effectively improving the rehabilitation treatment effect for patients.
[0051] The signal acquisition module includes an EEG signal acquisition module for acquiring EEG signals for motor imagery and a sensor for acquiring limb state information. Both the EEG signal acquisition module and the angular acceleration sensor are connected to the host computer.
[0052] The EEG signal acquisition module is an EEG cap:
[0053] The EEG cap collects the subject's motor imagery EEG signals in real time and transmits the collected information wirelessly to the host computer.
[0054] Angular acceleration sensors are fixed to the subject's limbs to record the subject's limb state information when electrical stimulation is output. The information is then transmitted in real time via Bluetooth and displayed on the system screen to assess the effectiveness of rehabilitation training.
[0055] The angular acceleration sensor is an angular acceleration actuator that records the angle of the subject's movement when electrical stimulation is output.
[0056] Joint locations include, but are not limited to, wrist joints and elbow joints;
[0057] The specific operation of the signal processing module is as follows:
[0058] Step 1: A Butterworth filter of 8-30Hz is used to filter out noise from the motor imagery EEG signal to obtain the effective signal;
[0059] Motor imagery EEG signals are primarily concentrated in specific frequency bands: beta waves (13–30 Hz) and alpha waves (8–12 Hz). The 8–30 Hz Butterworth filter has a flat passband without peaks or troughs, providing a flat frequency response. Using an 8–30 Hz Butterworth filter to remove noise from motor imagery EEG signals ensures integrity and accuracy.
[0060] Step 2: Use Independent Component Analysis (ICA) or wavelet transform to remove artifacts such as electrooculography (EOG) and electromyography (EMG) from the effective signal to obtain the true signal;
[0061] Electroencephalogram (EEG) signals are typically multichannel and a mixture of multiple sources of brain activity. Artifacts such as electrooculogram (EOG) and electromyogram (EMG) signals are usually mixed with noise caused by physiological activities. Using independent component analysis (ICA) or wavelet transform, it is possible to decompose multiple mixed signals into independent components, identify and separate artifacts such as EOG and EMG signals, selectively remove these artifacts, retain useful EEG signals, and obtain the true signal.
[0062] Step 3: Train the spatial domain feature vector of the real signal using the Co-space Pattern (CSP) method;
[0063] The Common Spatial Pattern (CSP) method calculates the covariance matrices for two imagined motion states. Then, eigenvalue decomposition is performed on the two covariance matrices to obtain eigenvalues and eigenvectors. Finally, a spatial filter is constructed to maximize the difference in variance between the two signal classes, thereby obtaining spatial domain eigenvectors with high discriminative power.
[0064] Step 4: Classify the spatial domain feature vectors using Support Vector Machine (SVM), Linear Discriminant Analysis (LDA), or Multilayer Perceptron (MLP);
[0065] Step 5: Evaluate the classification results using ROC curves, input the evaluation results into a real-time support vector machine (SVM) model to determine the motion intent, and output selection instructions based on the motion intent.
[0066] Choose the command to move left or remain still.
[0067] The leftward command is a confirmation command;
[0068] The resting instruction is an instruction to not make any decisions;
[0069] The eye-tracking module uses an eye tracker, which is fixed below the computer screen.
[0070] The eye-tracking module operates as follows:
[0071] Step 1: The eye tracker tracks the eye's gaze. The first selection interface displays the main category prompts. The first prompt is determined, and then it is determined whether the selection instruction is a leftward instruction. If it is, proceed to Step 2; if it is not, stop the operation.
[0072] Step 2: The display interface enters the second selection interface. The eye tracker tracks the eye's gaze and displays the detailed category prompts on the second selection interface to determine the second prompt instruction. Then, it determines whether the selection instruction is a leftward instruction. If the determination is yes, the second prompt instruction is converted into a treatment instruction and sent to the host computer. If the determination is no, the operation stops.
[0073] Specific Case 1:
[0074] Step 1: Eye tracker tracks eye gaze. The first selection interface displays main category prompts (…). Figure 1 (As shown) Identify the wrist joint of the first prompt command, then determine whether the selected command is a leftward command. If the determination is yes, proceed to step two; if the determination is no, stop the operation.
[0075] Step Two: The display interface will then proceed to the second selection screen. Figure 2 As shown), the eye tracker tracks eye gaze and displays detailed category prompts on the second selection interface to determine the second prompt instruction.
[0076] When the second prompt command is selected as wrist extension, and the command is selected as leftward, the treatment command is wrist extension.
[0077] When the second prompt command is selected as wrist extension and the command to remain at rest is selected, no treatment is performed.
[0078] When the second prompt command is selected as wrist flexion, and the command is selected as leftward, the treatment command is wrist flexion.
[0079] When the second prompt command is selected as wrist flexion and the command to remain at rest is selected, no treatment is performed.
[0080] When the second prompt command is selected, it returns to the previous screen; when the command is selected as the left command, the treatment command returns to the first selection screen.
[0081] When the second prompt command is selected to return to the previous screen, and the command to remain still is selected, no treatment will be performed.
[0082] Specific Case 2:
[0083] Step 1: Eye tracker tracks eye gaze. The first selection interface displays main category prompts (…). Figure 1 (As shown) Identify the elbow joint of the first prompt command, then determine whether the selected command is a leftward command. If the determination is yes, proceed to step two; if the determination is no, stop the operation.
[0084] Step Two: The display interface will then proceed to the second selection screen. Figure 3 As shown), the eye tracker tracks eye gaze and displays detailed category prompts on the second selection interface to determine the second prompt instruction.
[0085] When the second prompt command is selected as elbow extension, and the command is selected as leftward, the treatment command is elbow extension.
[0086] When the second prompt command is elbow extension and the command to remain at rest is selected, no treatment is given.
[0087] When the second prompt command is selected as elbow flexion, and the command is selected as leftward, the treatment command is elbow flexion.
[0088] When the second prompt command is elbow flexion and the command to remain at rest is selected, no treatment is given.
[0089] When the second prompt command is selected, it returns to the previous screen; when the command is selected as the left command, the treatment command returns to the first selection screen.
[0090] When the second prompt command is selected to return to the previous screen, and the command to remain still is selected, no treatment will be performed.
[0091] Specific Case 3:
[0092] Step 1: Eye tracker tracks eye gaze. The first selection interface displays main category prompts (…). Figure 1 (As shown) Identify the wrist + elbow joint of the first prompt command, then determine whether the selected command is a leftward command. If the determination is yes, proceed to step two; if the determination is no, stop the operation.
[0093] Step Two: The display interface will then proceed to the second selection screen. Figure 4 As shown), the eye tracker tracks eye gaze and displays detailed category prompts on the second selection interface to determine the second prompt instruction;
[0094] When the second prompt command is selected as wrist + elbow extension, and the command is selected as leftward, the treatment command is wrist + elbow extension.
[0095] When the second prompt command is selected as wrist and elbow extension, and the command to remain at rest is selected, no treatment is performed.
[0096] When the second prompt command is selected as wrist + elbow flexion, and the command is selected as leftward, the treatment command is wrist + elbow flexion.
[0097] When the second prompt command is selected as wrist + elbow flexion, and the command to remain at rest is selected, no treatment is performed.
[0098] When the second prompt command is selected, it returns to the previous screen; when the command is selected as the left command, the treatment command returns to the first selection screen.
[0099] When the second prompt command is selected to return to the previous screen, and the command to remain still is selected, no treatment will be performed.
[0100] The eye-tracking system uses infrared technology to determine the Pulcim spot formed on the corneal surface when the eye fixates on an object. It calibrates and fits the vector between the relative position of the spot and the center of the pupil with the position of the stimulus target, and correlates the subject's eye movement data with the target position on the screen to determine the position of the fixation point as a circle with a diameter of 96 pixels.
[0101] The electrical stimulator module includes a main control chip, a power supply module, and a constant current source module. The main control chip is electrically connected to the power supply module and the constant current source module.
[0102] The main control chip is STM32F407.
[0103] The H-bridge circuit in the constant current source module is used for stimulation output, generating stimulation pulses with alternating positive and negative polarities and adjustable amplitude and pulse width. The stimulation frequency can be adjusted autonomously according to the different needs of the subject, ensuring the targeted and effective nature of rehabilitation training.
[0104] Example 2: An application method of a rehabilitation system based on motor imagery EEG signals and eye tracking, the specific operation is as follows:
[0105] Step 1: Set the initial electrical stimulator current parameters and limb initial state information parameters of the electrical stimulator module;
[0106] Step 2: The subject sits in front of the display module and wears the EEG signal acquisition module. The EEG signal acquisition module collects motor imagery EEG signals from the parietal lobe related to the sensorimotor cortex. The patch electrodes connected to the output of the electrical stimulator module are attached to the treatment site.
[0107] Step 3: The signal processing module and the eye-tracking module transmit the treatment instructions to the host computer;
[0108] Step 4: The host computer of the angular acceleration sensor transmits the treatment command to the electrical stimulator module, the electrical stimulator module performs electrical stimulation treatment, and the angular acceleration sensor collects the subject's limb movement status information;
[0109] Step 5: Calculate the difference between the actual limb movement state information parameters and the limb initial state information parameters. If the difference is ≥0, the stimulation ends, and the initial electrical stimulator current parameters and limb initial state parameters are adjusted. If the difference is <0, treatment continues.
[0110] When the difference is ≥0, it indicates that the current treatment is in line with the expected effect, and the initial electrical stimulator current parameters and limb initial state parameters can be adjusted according to the treatment plan.
[0111] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will 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.
Claims
1. A rehabilitation system based on motor imagery EEG signals and eye tracking, characterized in that, include: Signal acquisition module: used to acquire EEG signals for motor imagery and record limb state information; Signal processing module: used to store the collected motor imagery EEG signals, determine the motor intention through the motor imagery EEG signals, and output selection instructions; Eye-tracking module: Used to track eye gaze and interface prompts, and coordinate with selection commands to issue treatment instructions; Electrical stimulator module: used to receive treatment commands and output stimulation pulse current according to the treatment commands; The host computer module is electrically connected to the signal acquisition module, eye tracking module, signal processing module, and electrical stimulator module. It is used to transmit the motor imagery EEG signals acquired by the signal acquisition module and the limb state information of the corresponding joints to the signal processing module, to transmit the selection instructions processed by the signal processing module to the eye tracking module, and to transmit the treatment instructions of the eye tracking module to the functional electrical stimulator module. Display module: Used to display the selection interface and limb status information. The display module is connected to the host computer.
2. The rehabilitation system based on motor imagery EEG signals and eye tracking according to claim 1, characterized in that, The signal acquisition module includes an EEG signal acquisition module for acquiring EEG signals for motor imagery and a sensor for acquiring limb state information. Both the EEG signal acquisition module and the angular acceleration sensor are connected to the host computer.
3. The rehabilitation system based on motor imagery EEG signals and eye tracking according to claim 1, characterized in that, The specific operation of the signal processing module is as follows: Step 1: A Butterworth filter of 8-30Hz is used to filter out noise from the motor imagery EEG signal to obtain the effective signal; Step 2: Use Independent Component Analysis (ICA) or wavelet transform to remove artifacts such as electrooculography (EOG) and electromyography (EMG) from the effective signal to obtain the true signal; Step 3: Train the spatial domain feature vector of the real signal using the Co-space Pattern (CSP) method; Step 4: Classify the spatial domain feature vectors using Support Vector Machine (SVM), Linear Discriminant Analysis (LDA), or Multilayer Perceptron (MLP); Step 5: The classification results are evaluated using ROC curves. The evaluation results are then fed into a real-time support vector machine (SVM) model to determine the motion intent. Based on the motion intent, selection instructions are output.
4. The rehabilitation system based on motor imagery EEG signals and eye tracking according to claim 3, characterized in that, Choose the command to move left or remain still.
5. The rehabilitation system based on motor imagery EEG signals and eye tracking according to claim 4, characterized in that, The eye-tracking module uses an eye tracker, which is fixed below the computer screen.
6. The rehabilitation system based on motor imagery EEG signals and eye tracking according to claim 5, characterized in that, The eye-tracking module operates as follows: Step 1: The eye tracker tracks the eye's gaze. The first selection interface displays the main category prompts. The first prompt is determined, and then it is determined whether the selection instruction is a leftward instruction. If it is, proceed to Step 2; if it is not, stop the operation. Step 2: The display interface enters the second selection interface. The eye tracker tracks the eye's gaze and displays the detailed category prompts on the second selection interface to determine the second prompt instruction. Then, it determines whether the selection instruction is a leftward instruction. If the determination is yes, the second prompt instruction is converted into a treatment instruction and sent to the host computer. If the determination is no, the operation stops.
7. The rehabilitation system based on motor imagery EEG signals and eye tracking according to claim 6, characterized in that, The electrical stimulator module includes a main control chip, a power supply module, and a constant current source module. The main control chip is electrically connected to the power supply module and the constant current source module.
8. A method for applying the rehabilitation system based on motor imagery EEG signals and eye tracking according to claim 7, characterized in that: The specific steps are as follows: Step 1: Set the initial electrical stimulator current parameters and limb initial state information parameters of the electrical stimulator module; Step 2: The subject sits in front of the display module and wears the EEG signal acquisition module. The EEG signal acquisition module collects motor imagery EEG signals from the parietal lobe related to the sensorimotor cortex and attaches the patch electrodes connected to the output of the electrical stimulator module to the treatment site. Step 3: The signal processing module and the eye-tracking module transmit the treatment instructions to the host computer; Step 4: The host computer of the angular acceleration sensor transmits the treatment command to the electrical stimulator module, the electrical stimulator module performs electrical stimulation treatment, and the angular acceleration sensor collects the subject's limb movement status information; Step 5: Calculate the difference between the actual limb movement state information parameters and the limb initial state information parameters. If the difference is ≥0, the stimulation ends, and the initial electrical stimulator current parameters and limb initial state parameters are adjusted. If the difference is <0, treatment continues.