Tail end traction type digital intelligent upper limb rehabilitation robot training system based on electrophysiological signals

Abstract

The invention discloses a tail end traction type digital intelligent upper limb rehabilitation robot training system based on electrophysiological signals, which is characterized in that firstly, the muscle force and the joint motion range of a patient are evaluated, and a personalized rehabilitation training plan is made for the patient according to an evaluation result; the plans are converted into codes which can be recognized by the robot on the computer, so that the patient is driven to train. Training modes comprise passive training and impedance training, the system monitors and analyzes electroencephalogram signals and force sensor data of a patient in real time, motion parameters of the robot can be dynamically adjusted according to the muscle strength level, the joint motion range and the electroencephalogram signals of the patient, and efficient and personalized upper limb rehabilitation training is achieved. The system fully considers individual differences of patients, has high flexibility and adaptability, and remarkably improves the upper limb rehabilitation training effect. Through fusion analysis of electrophysiological signals and application of a tail end traction technology, the system provides an advanced rehabilitation solution for patients with upper limb dysfunction.

Description

Technical Field

[0001] The present invention relates to the technical field of medical rehabilitation systems, specifically to an end-of-line, digitally intelligent upper limb rehabilitation robot training system based on electrophysiological signals. Furthermore, the present invention also relates to the fields of biomedical engineering and robotics, specifically including methods and techniques for motor dysfunction rehabilitation training using electrophysiological signals, as well as the design and application of a six-degree-of-freedom robotic system for rehabilitation medicine. Background Art

[0002] With the increasing aging of society and the accelerated pace of life, the incidence of upper limb motor dysfunction is gradually increasing. This dysfunction may be caused by stroke, trauma, neurological disease, or other reasons, seriously affecting patients' quality of life and work ability. Therefore, upper limb rehabilitation training is particularly important in the field of medical rehabilitation.

[0003] Traditional upper limb rehabilitation training methods typically rely on direct guidance from rehabilitation therapists, which has several limitations. First, due to limited therapist resources, patients often lack access to adequate personalized training time and intensity. Second, traditional training methods lack precise quantitative indicators, making it difficult to objectively assess and adjust a patient's recovery progress.

[0004] In recent years, the development of rehabilitation robots has made it possible to automate and personalize upper limb rehabilitation training. Existing rehabilitation robots can provide repeatable and controllable training, helping patients perform effective rehabilitation exercises in a safe environment. However, these robotic systems often lack real-time monitoring of the patient's physiological state, making it impossible to dynamically adjust training programs based on the patient's actual responses and needs.

[0005] Electrophysiological signals, particularly EEG signals, are important indicators of brain activity. They can reveal a patient's motor intentions and muscle activity, providing direct physiological feedback for rehabilitation training. Currently, research on the application of electrophysiological signals in rehabilitation robot motion planning is insufficient, and there is a lack of rehabilitation robot systems on the market that can adjust training plans based on patients' real-time electrophysiological signals.

[0006] Therefore, the present invention aims to provide an end-traction digital upper limb rehabilitation robot training system based on electrophysiological signals. The system can collect the patient's EEG signals in real time, and dynamically adjust the robot's motion planning based on signal analysis and evaluation of the patient's rehabilitation training effect, so as to achieve more personalized and effective upper limb rehabilitation training. Summary of the Invention

[0007] In order to solve the problem that the rehabilitation robot system in the prior art lacks real-time physiological status monitoring of patients and dynamic adjustment of training programs, the present invention proposes an end-traction digital upper limb rehabilitation robot training system based on electrophysiological signals.

[0008] In order to achieve the above object, the technical solution adopted by the present invention is:

[0009] A terminal traction-type digital upper limb rehabilitation robot training system based on electrophysiological signals includes: a chair, an EEG cap, a computer and a robot; the patient wears the EEG cap and sits on the chair, with his hands fixed to the traction rod of the robot's terminal joint using rehabilitation training fixation gloves. The robot is then controlled through motion trajectory coding to lead the patient's upper limbs to perform rehabilitation exercises such as flexion, extension, and rotation, and the EEG signals are analyzed in real time to adjust the training plan.

[0010] As a preferred technical solution of the present invention, in the upper limb rehabilitation robot training system:

[0011] The robot adopts the STR400 six-degree-of-freedom robot, and its end is equipped with an end traction device, which includes a fixed connector, a traction rod and a baffle. The fixed connector is used to connect the robot's end joint with the fixed traction rod. The traction rod provides a fixed grasping part for the patient's hand. The baffle is connected to the traction rod to prevent the patient's palm from slipping off the traction rod during rehabilitation training.

[0012] The EEG cap is worn on the patient's head, and fixing straps are fixed on the patient's forearms and upper arms. Elastic hook straps and spring dynamometers are added between the fixing straps. The patient wears rehabilitation training fixing gloves on his hands and fixes them on the end traction device of the STR400 six-degree-of-freedom robot.

[0013] The signal lines on the EEG cap are connected to the central area (Cz, C3, C4), frontal lobe (Fz, F3, F4), parietal lobe area (Pz, P3, P4) and visual field and motor imagery (O1, O2, P7, P8) channel ports on the amplifier, and the amplifier, EEG acquisition instrument host and computer are connected in sequence through connecting lines.

[0014] The computer not only has robot motion trajectory control software (Python) and EEG signal analysis software (Matlab), but also str-studio-win software and the STR studio V0.2.53 main interface; the robot's motion process can be seen in the str-studio-win software interface. If there is a problem with the programmed motion trajectory, the software interface will report an error, making it easy to modify the motion trajectory; the STR studio V0.2.53 main interface contains the robot's 3D model, emergency stop button and the robot's current posture parameters.

[0015] The present invention also proposes a method for performing personalized training using the training system, which comprises the following steps:

[0016] First, the patient's muscle strength and joint mobility are assessed, and then a personalized rehabilitation training plan is developed based on the results;

[0017] The training plan is then converted into robot-recognizable code on a computer to drive the patient to perform the training. The training modes include passive training and resistance training. During training, the patient's EEG signals and force sensor data are monitored and analyzed in real time to dynamically adjust the robot's motion parameters, ensuring personalized and efficient rehabilitation training.

[0018] Finally, after a course of treatment, a rehabilitation effect evaluation is conducted. Through the evaluation results, patients and doctors can understand the treatment effect more clearly and provide a basis for formulating the next course of treatment plan.

[0019] Compared with the prior art, the present invention has the following beneficial effects:

[0020] 1. Personalized treatment: Based on the patient's information and EEG signal changes during training, a personalized rehabilitation training plan can be customized to better meet the rehabilitation needs of each patient. At the same time, it can significantly improve the accuracy of individual rehabilitation and accelerate the patient's recovery process to a certain extent.

[0021] 2. Real-time feedback and adjustment: Real-time monitoring of the patient's EEG activity allows for immediate adjustments to the robot's movements to match the patient's intentions and physiological state, improving the efficiency and safety of training.

[0022] 3. Enhance patient participation: By responding to the patient's EEG signals, the patient can participate more actively in the rehabilitation process, improving the patient's enthusiasm and motivation for rehabilitation.

[0023] 4. Objective evaluation: EEG signals can provide objective data to evaluate the patient's recovery progress and reduce the uncertainty of subjective judgment.

[0024] 5. Improve training effect: By precisely controlling the movement of the robot, the accuracy and repeatability of the training movements can be ensured, thereby improving the effect of rehabilitation training.

[0025] 6. Reduce physical burden: Robots can take on part or all of the training work, reducing the workload of therapists and reducing the risk of injury to patients due to lack of strength or poor coordination.

[0026] 7. Data recording and analysis: The system can record large amounts of training data for subsequent analysis, helping researchers and clinicians better understand the rehabilitation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a structural schematic diagram of the end-traction digital upper limb rehabilitation robot training system based on electrophysiological signals of the present invention.

[0028] Figure 2 This is a schematic diagram of the structure of the STR400 six-degree-of-freedom robot.

[0029] Figure 3 It is a structural schematic diagram of the terminal traction device of the present invention.

[0030] Figure 4 This is a schematic diagram of EEG signal acquisition.

[0031] Figure 5 It is a rehabilitation training flow chart.

[0032] Figure 6 It is a flowchart for rehabilitation program design.

[0033] Figure 7 This is a flowchart for muscle strength assessment.

[0034] Figure 8 This is a flowchart for joint range of motion assessment.

[0035] Figure 9 It is a flow chart of the robot feedback mechanism.

[0036] Figure 10 It is a flowchart of rehabilitation effect evaluation.

[0037] Figure 1 、 3 、4 The meanings of the reference numerals are:

[0038] 1-patient, 2-chair, 3-EEG cap, 4-computer, 5-robot, 6-terminal traction device, 7-rehabilitation training fixation gloves, 8-robot support, 9-robot data transmission cable, 10-fixed connector, 11-traction rod, 12-baffle, 18-EEG acquisition device host, 19-EEG signal amplifier, 20-EEG data transmission cable, 21-professional therapist, 22-fixation belt, 23-elastic hook belt. DETAILED DESCRIPTION

[0039] Example 1

[0040] See also Figure 1As shown, this embodiment provides an end-traction digital upper limb rehabilitation robot training system based on electrophysiological signals, which mainly includes: a chair 2, an EEG cap 3, a computer 4, and a robot 5. Patient 1 wears the EEG cap 3 and sits on the chair 2 in a comfortable posture. The hand is fixed to the traction rod 11 of the robot end-traction device with rehabilitation training gloves 7. Then, the robot 5 is controlled by motion trajectory coding to carry out rehabilitation exercises such as flexion, extension, and rotation of the patient's upper limb, and the EEG signals are analyzed in real time to adjust the training plan. The following describes the composition of each component:

[0041] 1. The chair used for rehabilitation training must be able to withstand a load of at least 150 kg; the chair height must be adjustable so that the patient can place both feet flat on the ground when sitting with the knees bent at 90 degrees; the armrests of the chair must also be automatically adjustable to help the patient maintain balance when standing up, sitting down or performing upper limb rehabilitation exercises; the chair should have a reliable locking mechanism to ensure that the chair does not slide during rehabilitation exercises.

[0042] 2. The robot 5 used in upper limb rehabilitation training is a STR400 six-degree-of-freedom robot (such as Figure 2 ), has the following characteristics:

[0043] ① High flexibility: The STR400 robot is a small six-axis robotic arm with an all-carbon fiber body. It uses six high-performance robotic joint modules to simulate complex movements of the upper limbs, providing patients with multi-directional and multi-angle training, which helps to fully restore upper limb function.

[0044] ②Precise control: The STR400 robotic platform is equipped with a high-precision control system that can accurately control motion trajectory and force, which is crucial for rehabilitation training and can ensure the safety and effectiveness of training.

[0045] ③Repeatability: Robotic rehabilitation equipment can provide consistent and repeatable training, which is very important for the rehabilitation process and helps patients practice according to the established rehabilitation plan.

[0046] ④ Adaptability: The robot’s motion parameters, such as speed, range, and force, can be adjusted according to the rehabilitation needs and progress of different patients to achieve personalized rehabilitation training.

[0047] ⑤Safety: The STR400 robot control platform is equipped with an emergency stop button. These safety features can provide protection for patients and reduce accidental injuries during training.

[0048] The rehabilitation robot is fixed on a robot support 8, which has wheels at the bottom for easy movement and a lock switch to prevent the robot from moving under force during movement. It also has a height adjustment function.

[0049] like Figure 3 As shown, the end of the rehabilitation robot is equipped with an end traction device 6, which includes the following components: ① Fixed connecting head 10: used to connect the end joint of the robot and the fixed traction rod 11; ② Traction rod 11: The fixed grasping part of the patient's hand, which is designed to have an appropriate diameter and a moderate length to facilitate the patient's grasping, and its shape is L-shaped, which can meet the grasping needs in different directions; ③ Baffle 12: Connected to the traction rod, it can effectively prevent the patient's palm from slipping off the traction rod during rehabilitation training.

[0050] 3. The patient sits on a height-adjustable chair 2. After adjusting the chair to the appropriate height, the patient is fitted with an EEG cap 3 and secured with securing straps 22 around their forearms and upper arms. Elastic hook straps 23 and a spring dynamometer are placed between the securing straps. Next, the patient's hands are fitted with rehabilitation training gloves 7 and secured to the end-of-line traction device 6 of the STR400 six-degree-of-freedom robot 5. The robot's rehabilitation training trajectory, pre-programmed on computer 4, is then executed. By inputting commands via data transmission line 9, the six-degree-of-freedom robot 5 can then guide the patient through upper limb rehabilitation training.

[0051] 4. Rehabilitation training immobilization gloves should have the following features: Made of composite fabric, they are soft, strong, breathable, and comfortable. Extended straps can be adjusted and tightened to provide more stable hand fixation. A finger-separated design effectively prevents spasms and slippage. Adjustable fixing structures can accommodate different hand sizes.

[0052] 5. If Figure 4 As shown, the specific implementation steps for collecting EEG signals include: first, the EEG acquisition instrument host 18 is connected to the power supply and connected to the amplifier 19 with a connecting line 20, and then the signal line on the EEG cap 3 is connected to the central area (Cz, C3, C4), frontal lobe (Fz, F3, F4), parietal lobe area (Pz, P3, P4) and visual field and motor imagery (O1, O2, P7, P8) channel ports on the amplifier 19, then the patient is correctly fitted with the EEG cap, and conductive paste is applied to the electrodes of the relevant channels to ensure the connection quality of each channel, and finally the patient's EEG signals can be collected.

[0053] 6. The computer contains not only the robot motion trajectory control software (Python) and EEG signal analysis software (Matlab), but also the str-studio-win software and the STR studio V0.2.53 main interface. The str-studio-win interface displays the robot's motion progress. If there are any issues with the programmed motion trajectory, the software will report an error, making it easy to modify the trajectory. The STR studio V0.2.53 main interface includes the robot's 3D model, an emergency stop button, and the robot's current posture parameters.

[0054] The specific working principle of this invention is as follows: First, the patient's muscle strength and joint mobility are assessed, and then a professional physician develops a personalized rehabilitation training plan based on the results. This training plan is then converted into a code recognizable by the robot on a computer, which drives the patient to perform the training. The training modes include passive training and impedance training. During training, the patient's EEG signals and force sensor data are monitored and analyzed in real time, thereby dynamically adjusting the robot's motion parameters to ensure personalized and efficient rehabilitation training. After a course of treatment, the rehabilitation effect is evaluated. The evaluation results provide patients and physicians with a clearer understanding of the treatment effects, providing a basis for formulating the next course of treatment.

[0055] Example 2

[0056] This embodiment proposes a method for performing personalized training based on the upper limb rehabilitation robot training system. Figure 5 As shown, the specific steps include:

[0057] 1. Conduct a comprehensive assessment of the patient's condition and tailor a personalized rehabilitation plan based on the assessment results.

[0058] like Figure 6 As shown, the details of the condition assessment are:

[0059] ①Basic information of the patient: name, age, gender, etc.

[0060] ② Medical history information: Ask the patient about their medical history in detail, including the cause of the disease, course of the disease, previous treatment, medication use, etc.

[0061] ③ Assessment of current physical condition: The patient's muscle strength and joint mobility are assessed using a portable digital muscle strength tester (MicroFET2) and a wearable joint mobility measuring instrument (Qindong NEO-SMART).

[0062] like Figure 7 As shown, the patient's muscle strength assessment includes the following tests using a portable digital muscle tester (MicroFET2):

[0063] Grip strength test, shoulder flexion and extension strength test, shoulder abduction and adduction strength test, elbow flexion and extension strength test, forearm pronation and supination strength test, and wrist flexion and extension strength test. Based on the observed muscle contraction, joint movement, and measured muscle strength, the patient's muscle strength level can be divided into 0 to 5 levels:

[0064] Level 0: Zero (0); no muscle contraction, no muscle strength at all; equivalent to 0% of normal muscle strength.

[0065] Grade 1: Trace (T); slight muscle contraction but no joint movement; equivalent to 10% of normal muscle strength.

[0066] Grade 2: Poor (P); joint movement is possible without gravity, equivalent to 25% of normal muscle strength.

[0067] Level 3: Fair (F); under normal gravity, there is joint movement but cannot resist resistance; equivalent to 50% of normal muscle strength.

[0068] Grade 4: Good (G); able to resist light resistance but not moderate resistance; equivalent to 75% of normal muscle strength.

[0069] Level 5: Normal (N); able to resist moderate resistance, normal muscle strength; equivalent to 100% of normal muscle strength.

[0070] like Figure 8 As shown, the patient's joint range of motion assessment mainly includes the following steps:

[0071] A wearable range of motion meter (Qindong NEO-SMART) was placed 5 cm above the patient's elbow and wrist to record the range of motion of the shoulder, elbow, forearm, and wrist. This included the extremes of shoulder flexion, extension, abduction, adduction, and internal and external rotation; elbow flexion and extension; forearm pronation and supination; and wrist flexion, extension, ulnar deviation, and radial deviation. Before and after training, the range of motion of each joint was recorded using bar graphs and line charts to visually demonstrate changes in joint motion before and after training.

[0072] 2. Based on the similarity of their conditions, the patients were divided into two groups: the experimental group was trained using the rehabilitation robot system developed in this study, while the control group was trained using traditional rehabilitation training methods. Rehabilitation plan for the experimental group was designed as follows:

[0073] ① Rehabilitation goal setting: Set practical short-term and long-term rehabilitation goals based on the patient's specific situation.

[0074] ② Develop a rehabilitation plan: Based on the patient's condition, physical examination report and rehabilitation goals, professional therapists will develop personalized rehabilitation training methods and content, including training methods, frequency, intensity, etc.

[0075] ③ Rehabilitation trajectory planning: Edit the rehabilitation training movements in the rehabilitation training plan into code, convert it into a language that the robot can recognize, and let the robot lead the patient in exercise.

[0076] The robot's motion planning steps are as follows: The personalized rehabilitation training methods and content developed by a professional therapist are converted into code that can be understood by the robot 5 using Python software on a computer 4, enabling it to simulate normal physiological movements, such as flexing, extending, and rotating the patient's shoulder, elbow, or wrist joints, ensuring a smooth and continuous trajectory. After the design is completed, the planned motion is simulated using simulation software to check for collisions, overruns, and other issues. The motion plan is then optimized based on the simulation results. After ensuring safety and comfort, a small-scale clinical trial can be conducted to collect patient feedback and adjust the motion plan accordingly.

[0077] 3. Implementation of rehabilitation training

[0078] For the experimental group, the detailed training steps are as follows: First, the patient sits on a chair with adjustable height. After adjusting the seat height to the appropriate position, the patient is given an EEG cap, and the fixing straps 22 are fixed on the patient's forearms and upper arms, and elastic hook straps 23 and spring dynamometers are added between the fixing straps. Next, the patient's hands are given rehabilitation training fixing gloves 7, which are fixed to the end traction device 6 of the STR400 six-degree-of-freedom robot 5. Subsequently, the robot rehabilitation training motion trajectory preset in the computer 4 is run. By inputting instructions through the data transmission line 9, the six-degree-of-freedom robot 5 can drive the patient to perform upper limb rehabilitation training. At the same time, the robot's motion plan is adjusted according to the analysis of the patient's muscle strength, joint mobility and EEG signals.

[0079] Rehabilitation exercises include two modes: passive exercise and resistance exercise. Passive exercise eliminates the need for elastic hook straps 23 between the upper and lower arm straps 22. Resistance exercise involves adding appropriate elastic hook straps 23 based on the patient's muscle strength level to increase resistance during arm movement. During rehabilitation exercises, patients also need to attach a force sensor—a spring dynamometer—between the straps 22 to ensure the force they experience during exercise is within an acceptable range to avoid secondary injury.

[0080] When wearing an EEG cap to measure EEG signals, the main EEG signal channels measured are the central area (Cz, C3, C4), frontal lobe (Fz, F3, F4), parietal area (Pz, P3, P4) and visual field and motor imagery (O1, O2, P7, P8). The changes in the EEG signals of the above channels can better evaluate the effect of rehabilitation training and observe the process of neural remodeling. During the training process, the EEG cap monitors and records the patient's EEG signals in real time. The computer analyzes the signals, mainly including the analysis of changes in the frequency domain alpha waves (8-12Hz) and beta waves (12-30Hz) to determine the patient's training situation, and adjusts the motion parameters of the six-degree-of-freedom robot in real time according to the patient's muscle strength, joint mobility and EEG analysis results, thereby ensuring the personalization and efficiency of rehabilitation training.

[0081] like Figure 9 As shown in the figure, the robot's motion planning is adjusted based on analysis of the patient's muscle strength, joint range of motion, and EEG signals. The main adjustments are as follows: ① Force adjustment: Dynamically adjusts the robot's traction force based on the patient's muscle strength level. ② Range of motion adjustment: Adjusts the robot's range of motion based on the patient's joint range of motion to avoid causing pain or injury. ③ Intensity adjustment: If the EEG signals indicate that the patient is comfortable and wishes to increase force, the robot's movement speed, acceleration, and other parameters are increased. If the patient feels fatigued or uncomfortable and wishes to reduce force, the movement speed, acceleration, and other parameters are decreased.

[0082] 4. Rehabilitation effect evaluation

[0083] Through the above steps, the patients received a complete course of training. After the training, the patients’ condition and rehabilitation effects before and after the training were compared in detail, and the rehabilitation effects of the experimental and control groups were comprehensively evaluated. This process effectively quantified and analyzed the rehabilitation progress of each patient, thus providing a scientific basis and optimized guidance for future rehabilitation treatment (such as Figure 10 shown).

[0084] After one course of treatment, the effects of the two groups of patients were evaluated respectively.

[0085] ① Quantitative evaluation: Measure the patient's upper limb muscle strength and joint mobility again, and judge the treatment effect by comparing the changes in muscle strength level and joint mobility before and after treatment.

[0086] ② Qualitative evaluation: Compare the changes in EEG signals before and after rehabilitation training to objectively evaluate the rehabilitation effect.

[0087] The specific implementation steps for the patient's EEG signal processing and analysis are as follows: The collected EEG signals are analyzed using the computer software Matlab, mainly including the analysis of the frequency domain α wave (8-12Hz) and β wave (12-30Hz) waveform changes, among which:

[0088] Alpha wave activity is related to relaxation, concentration, and consciousness. During rehabilitation training, analyzing changes in a patient's alpha wave waveform can provide information about the patient's degree of relaxation, concentration, and cognitive load. The methods for adjusting upper limb rehabilitation training based on changes in the alpha wave waveform are as follows: ① Monitor the state of relaxation: If alpha wave activity increases, it indicates that the patient is in a more relaxed state. The training difficulty can be appropriately increased, and the patient can be encouraged to perform more active movements. If alpha wave activity decreases, it may indicate that the patient is nervous or overexerting. The training difficulty can be reduced, and the patient can be guided to perform deep breathing and relaxation exercises. ② Adjust the training rhythm: When alpha wave activity is stable, the current training rhythm can be maintained. If alpha wave activity fluctuates, the training rhythm may need to be adjusted, giving the patient more rest time or changing the training rate. Beta waves are related to active movement, concentration, and cognitive processing.

[0089] During rehabilitation training, analyzing changes in the patient's beta wave waveform can help understand the patient's cognitive and motor activity level. The following are methods for adjusting upper limb rehabilitation training based on changes in the beta wave waveform: ① Monitor attention level: If beta wave activity increases, it indicates that the patient is in a state of high attention and alertness, which may be a good time to perform complex or challenging training; if beta wave activity decreases, it may indicate that the patient's attention has decreased, and it is necessary to simplify the training task or provide additional incentives to improve attention. ② Adjust the difficulty of training: When beta wave activity is stable and the intensity is moderate, the current training difficulty can be maintained; if beta wave activity is too high, it may indicate that the patient is overworking, and the training difficulty should be appropriately reduced to avoid excessive fatigue; if beta wave activity is too low, it may be necessary to increase the challenge of the training task to improve the patient's participation and attention.

[0090] The quantitative evaluation method for the patient's rehabilitation effect after training is:

[0091] Compare the changes in the patient's muscle strength level before and after training and the changes in the bar and line graphs of each joint angle to determine the rehabilitation effect. If the patient's muscle strength level increases and some joint angles increase, the rehabilitation training is more effective.

[0092] The qualitative evaluation method for the rehabilitation effect of patients after training is:

[0093] The rehabilitation effect is objectively evaluated by comparing the changes in EEG signals before and after rehabilitation training. The EEG signal parameters compared include changes in frequency band power, EEG signal entropy, and EEG signal coherence:

[0094] ①Band Power

[0095] Frequency band power refers to the strength of EEG signals within a specific frequency range. Comparing changes in alpha and beta band power before and after training, increases in alpha and beta band power are generally associated with improved motor function, indicating increased brain efficiency in motor planning and execution.

[0096] ② Entropy of EEG signals

[0097] Entropy is a measure of the complexity and randomness of EEG signals. Higher entropy values ​​indicate a more complex and random signal. Comparing entropy values ​​before and after training shows that increased entropy is generally associated with improved motor function, indicating enhanced brain processing capabilities for motor control and sensory feedback.

[0098] ③ Coherence of EEG signals

[0099] Coherence is a measure of the synchronization of EEG signals between different brain regions. High coherence indicates increased synchronization between these regions. Comparing changes in EEG coherence before and after training reveals that increased coherence is generally associated with improved motor function, indicating more efficient information transfer between brain regions.

[0100] The above content is merely an example and explanation of the concept of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the claims, they should all fall within the scope of protection of the present invention.

Claims

1. An end-traction digital upper limb rehabilitation robot training system based on electrophysiological signals, characterized in that: include: A chair (2), an EEG cap (3), a computer (4) and a robot (5); a patient (1) wears an EEG cap (3), sits on a chair (2), and fixes his hands on a traction rod (11) of the end joint of the robot (5) using a rehabilitation training fixation glove (7). Then, the robot (5) is controlled through motion trajectory coding to carry out rehabilitation exercises such as flexion, extension, and rotation of the patient's upper limbs, and the EEG signal is analyzed in real time to adjust the training plan.

2. The training system according to claim 1, characterized in that The robot (5) is a STR400 six-degree-of-freedom robot, and a terminal traction device (6) is installed at the terminal end. The terminal traction device (6) includes a fixed connection head (10), a traction rod (11) and a baffle (12). The fixed connection head (10) is used to connect the robot terminal joint with the fixed traction rod (11). The traction rod (11) provides a fixed grasping position for the patient's hand. The baffle (12) is connected to the traction rod (11) to prevent the patient's palm from slipping off the traction rod during rehabilitation training.

3. The training system according to claim 1, characterized in that The electroencephalogram cap (3) is worn on the patient's head, and fixing straps (22) are fixed on the patient's forearms and upper arms. Elastic hook straps (23) and spring dynamometers are added between the fixing straps. The patient wears rehabilitation training fixing gloves (7) on his hands, and the gloves are fixed on the end traction device (6) of the STR400 six-degree-of-freedom robot (5).

4. The training system according to claim 1, characterized in that The signal line on the EEG cap (3) is connected to the central area, frontal lobe, parietal lobe area and visual field and motor imagery channel ports on the amplifier (19), and the amplifier (19), the EEG acquisition instrument host (18) and the computer (4) are connected in sequence via a connecting line (20).

5. The training system according to claim 1, characterized in that The computer (4) is provided with not only robot motion trajectory control software (Python) and EEG signal analysis software (Matlab), but also str-studio-win software and STR studio V0.2.53 main interface; the robot motion process can be seen in the str-studio-win software interface, and if there is a problem with the programmed motion trajectory, the software interface will report an error, making it easy to modify the motion trajectory; the STR studio V0.2.53 main interface contains the robot's 3D model, emergency stop button and the robot's current posture parameters.

6. A method for performing personalized training using the training system according to any one of claims 1 to 5, characterized in that: Here are the steps: First, the patient's muscle strength and joint mobility are assessed, and then a personalized rehabilitation training plan is developed based on the results; Then, the training plan is converted into robot-recognizable code on the computer to drive the patient to perform training. The training mode includes passive training and impedance training. During training, the patient's EEG signals and force sensor data are monitored and analyzed in real time, so as to dynamically adjust the robot's motion parameters to ensure the personalization and efficiency of rehabilitation training. Finally, after a course of treatment, a rehabilitation effect evaluation is conducted. Through the evaluation results, patients and doctors can have a clearer understanding of the treatment effect and provide a basis for formulating the next course of treatment plan.

Images