Non-invasive brain-computer interface rehabilitation robot

Abstract

The application discloses a non-invasive brain-computer interface rehabilitation robot and relates to the technical field of intelligent rehabilitation equipment, and solves the technical problem that the existing rehabilitation robot mainly adopts a set program to perform repetitive rehabilitation training, and patients are not actively involved, the passive training mode limits the autonomous control and participation of the patients, and the rehabilitation effect is limited, especially in the aspect of neural injury repair, and the effect is poor; the application obtains motion data of the patient; an artificial intelligence model is trained based on historical motion data to obtain a motion recognition model; the motion data is recognized based on the motion recognition model to obtain a recognition result; the patient is subjected to rehabilitation training through the rehabilitation robot based on the recognition result; a rehabilitation effect evaluation coefficient of the patient is calculated based on the electromyographic characteristic data; and the recovery progress of the patient is judged based on the rehabilitation effect evaluation coefficient and a preset rehabilitation effect evaluation threshold value, so that the above technical problem is solved.

Description

Technical Field

[0001] This invention belongs to the field of intelligent rehabilitation equipment, specifically a non-invasive brain-computer interface rehabilitation robot. Background Technology

[0002] Rehabilitation robots primarily passively guide patients through rehabilitation training. Representative examples include Hocoma's Lokomat and Germany's LokoHelp robot. Hocoma's Lokomat provides gait patterns consistent with human physiology, offering task-specific training. Based on continuous feedback and assessment, it gradually increases training volume and intensity, enhancing patient participation, improving gait training efficiency, and significantly accelerating the patient's rehabilitation process. Germany's LokoHelp lower limb rehabilitation robot training system adjusts gait and posture according to the individual patient's functional condition, guiding the patient's lower limbs on a treadmill, enabling extensive gait training.

[0003] Existing rehabilitation robots mainly use pre-set programs for repetitive rehabilitation training, in which patients are not actively involved. This passive training method limits the patient's autonomy and participation, resulting in limited rehabilitation efficacy, especially in the repair of nerve damage. At the same time, passive training can cause users to experience obvious pulling sensations, which affects the effectiveness of use.

[0004] Therefore, this invention proposes a non-invasive brain-computer interface rehabilitation robot to solve the above problems. Summary of the Invention

[0005] This invention aims to at least solve one of the technical problems existing in the prior art. To this end, this invention proposes a non-invasive brain-computer interface rehabilitation robot to solve the technical problem that existing rehabilitation robots mainly use pre-set programs for repetitive rehabilitation training, in which patients are not actively involved. This passive training method limits the patient's autonomous control and participation, resulting in limited rehabilitation efficacy, especially in the repair of nerve damage. At the same time, passive training can cause users to experience obvious pulling sensations, which affects the effectiveness of use.

[0006] To achieve the above objectives, a first aspect of the present invention provides a non-invasive brain-computer interface rehabilitation robot, comprising: a data acquisition module, a data analysis module, a rehabilitation training module, and a rehabilitation efficacy evaluation module;

[0007] Data acquisition module: used to acquire the patient's motion data; the motion data includes: electroencephalogram (EEG) characteristic data and electromyogram (EMG) characteristic data;

[0008] Data analysis module: Trains an artificial intelligence model based on historical motion data to obtain a motion recognition model; uses the motion recognition model to identify motion data and obtain recognition results;

[0009] Rehabilitation training module: Based on the recognition results, rehabilitation robots are used to provide rehabilitation training to patients;

[0010] The rehabilitation efficacy assessment module calculates the patient's rehabilitation efficacy assessment coefficient based on electromyographic characteristic data; and judges the patient's recovery progress based on the rehabilitation efficacy assessment coefficient and the preset rehabilitation efficacy assessment threshold.

[0011] Preferably, acquiring the patient's motion data includes:

[0012] The EEG cap collects real-time EEG signals from brain activity areas of patients during rehabilitation training using a rehabilitation robot.

[0013] Electromyography (EMG) signals of the affected limb are collected in real time by an EMG sensor attached to the skin surface of the muscles that move in the affected limb.

[0014] By performing time-domain and frequency-domain analyses on EEG and EMG signals, we obtained the EEG characteristic data corresponding to the EEG signals and the EMG characteristic data corresponding to the EMG signals.

[0015] It should be noted that brain activity areas refer to specific parts of the brain that are related to specific functions or tasks. For example, the motor cortex is responsible for controlling body movement, and the visual cortex processes visual information. By collecting the electroencephalogram (EEG) signals of these areas through an EEG cap, changes in electrical activity related to the patient's motor intentions can be captured.

[0016] Electroencephalogram (EEG) signals reflect the electrical activity of neurons in the brain. Specific EEG patterns correspond to different cognitive and motor intentions. By analyzing these patterns, the patient's motor intentions can be decoded, thereby achieving precise control of the rehabilitation robot.

[0017] Although the muscles of the affected limb of a paralyzed patient cannot contract involuntarily, some reflex activities and autonomic nervous system activities still exist; for example, when the patient tries to move the affected limb, some reflex muscle activities are triggered, which can be captured by electromyography sensors and used to help determine the patient's movement intentions.

[0018] The EEG characteristic data includes peak value, root mean square (RMS), zero crossover rate, signal amplitude, and signal power; the EMG characteristic data includes peak value, RMS, zero crossover rate, signal amplitude, and signal power.

[0019] Preferably, training the artificial intelligence model based on historical motion data includes:

[0020] A1: Extract historical EEG and electromyography (EMG) feature data from historical motion data;

[0021] A2: By obtaining the mapping relationship between historical EEG feature data and motor intention data, as well as the mapping relationship between historical EMG feature data and motor intention data from the database; where motor intention data includes: forward, backward, sprinting and jogging, etc.

[0022] A3: An artificial intelligence model is trained based on historical EEG feature data to obtain a motor intention recognition model; an artificial intelligence model is trained based on historical EMG feature data to obtain a motor intention verification model; wherein, the artificial intelligence model includes: convolutional neural network or deep belief network;

[0023] A4: The motion recognition model is composed of a motion intent recognition model and a motion intent verification model.

[0024] It should be noted that the historical motion data refers to the historical data of the collected motion data;

[0025] The database refers to existing public datasets that map the relationships between EEG feature data and motor intention data, as well as the mapping relationships between EMG feature data and motor intention data, including BCI Competition Datasets, PhysioNet, and OpenBMI.

[0026] Preferably, the step of training an artificial intelligence model based on historical EEG feature data to obtain a motion intention recognition model includes:

[0027] Historical EEG feature data is integrated into model input data, and the motion intention data corresponding to historical EEG feature data is integrated into model output data.

[0028] An artificial intelligence model is trained based on the model input data and model output data to obtain a motion intention recognition model.

[0029] Preferably, the step of training an artificial intelligence model based on historical electromyography feature data to obtain a movement intention verification model includes:

[0030] Historical electromyography (EMG) feature data is integrated into standard input data, and the motion intention data corresponding to historical EMG feature data is integrated into standard output data.

[0031] An artificial intelligence model is trained based on standard input and standard output data to obtain a motion intent verification model.

[0032] Preferably, the process of identifying motion data based on a motion recognition model includes:

[0033] B1: Extract real-time acquired EEG and EMG feature data;

[0034] B2: Input the EEG feature data into the motor intention recognition model to obtain the recognition result;

[0035] B3: Input the electromyographic feature data into the motion intention verification model to obtain the verification and recognition results;

[0036] B4: Determine whether the verification result is consistent with the recognition result; if yes, output the recognition result; if no, return to step B1 until the verification result is consistent with the recognition result.

[0037] It should be noted that the identification results, verification identification results, and motion intention data content attributes are consistent, that is, the identification results and verification identification results both include: forward, backward, fast running, and slow running, etc.

[0038] Preferably, the calculation of the patient's rehabilitation efficacy evaluation coefficient based on electromyographic characteristic data includes:

[0039] The duration of rehabilitation training for patients was obtained, and the root mean square value and zero crossover rate were extracted from the electromyography feature data.

[0040] The duration of rehabilitation training is denoted as T, the root mean square value as R, and the zero crossover rate as L.

[0041] Through the formula:

[0042] P=α / e^[(T-ZT)^2 / ZT]+β×e^[(R-CR) / ln(CR)]+ɡ×e^[(L-CL) / ln(CL)];

[0043] The rehabilitation efficacy evaluation coefficient is calculated; where P is the rehabilitation efficacy evaluation coefficient, α, β, and g are proportionality coefficients, and 0 < α < 1, 0 < β < 1, 0 < g < 1, ZT is the optimal rehabilitation training duration, CR is the initial root mean square value, and CL is the initial zero crossover rate.

[0044] It should be noted that the optimal duration of rehabilitation training is determined by experts in this field based on experience and the patient's actual condition; the initial root mean square value and the initial zero crossover rate refer to the root mean square value and zero crossover rate of the electromyographic signal before the patient has undergone rehabilitation training.

[0045] Training time that is too short may prevent patients from effectively improving their muscle endurance and overall physical fitness; training time that is too long may lead to excessive muscle fatigue, increasing the risk of injury, and long-term fatigue may also affect the muscle's recovery and regeneration capabilities.

[0046] An increase in the root mean square value relative to the initial root mean square value indicates enhanced muscle activation, which is a manifestation of muscle strength recovery. In rehabilitation training, this shows that the patient's muscle strength and activation ability are improving.

[0047] An increase in the zero-crossing rate relative to the initial zero-crossing rate indicates an increase in the frequency of muscle activation, which is a manifestation of the recovery of neural control ability. In rehabilitation training, this shows that the patient's muscle activation frequency and coordination are improving.

[0048] Preferably, the step of judging the patient's recovery progress based on the rehabilitation efficacy assessment coefficient and the preset rehabilitation efficacy assessment threshold includes:

[0049] Determine whether the rehabilitation efficacy assessment coefficient is greater than or equal to the preset rehabilitation efficacy assessment threshold; if yes, it indicates that the patient's rehabilitation efficacy is good; if no, it indicates that the patient's rehabilitation efficacy is poor.

[0050] It should be noted that the preset rehabilitation efficacy assessment threshold is set by experts in the field based on their experience and the patient's initial condition.

[0051] Compared with the prior art, the beneficial effects of the present invention are:

[0052] Existing rehabilitation robots primarily employ pre-programmed, repetitive rehabilitation training with patients not actively participating. This passive training method limits patient autonomy and participation, resulting in limited rehabilitation efficacy, particularly in nerve damage repair. Furthermore, passive training can cause a noticeable pulling sensation, negatively impacting usability. This invention addresses this issue by using a data acquisition module to acquire real-time EEG and EMG data from the patient. A data analysis module trains an artificial intelligence model to obtain a motor intention recognition and verification model, enabling accurate recognition and verification of the patient's motor intentions. The rehabilitation training module provides personalized training based on the recognition results, while the rehabilitation efficacy assessment module calculates an evaluation coefficient and uses preset thresholds to determine the patient's recovery progress. This approach not only improves patient participation and rehabilitation effectiveness but also ensures synchronization and coordination in human-computer interaction through real-time feedback and adaptive control algorithms, significantly enhancing the user experience and rehabilitation training results. This invention also addresses the problem of existing rehabilitation robots relying on pre-programmed, repetitive, passive training by introducing non-invasive brain-computer interface technology. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 This is a schematic diagram of a rehabilitation robot module according to an embodiment of the present invention;

[0055] Figure 2 This is a schematic diagram of the workflow of the rehabilitation robot according to an embodiment of the present invention. Detailed Implementation

[0056] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0057] Please see Figures 1-2 The first aspect of the present invention provides a non-invasive brain-computer interface rehabilitation robot, including: a data acquisition module, a data analysis module, a rehabilitation training module, and a rehabilitation efficacy evaluation module;

[0058] Data acquisition module: used to acquire the patient's motion data; the motion data includes: electroencephalogram (EEG) characteristic data and electromyogram (EMG) characteristic data;

[0059] Data analysis module: Trains an artificial intelligence model based on historical motion data to obtain a motion recognition model; uses the motion recognition model to identify motion data and obtain recognition results;

[0060] Rehabilitation training module: Based on the recognition results, rehabilitation robots are used to provide rehabilitation training to patients;

[0061] The rehabilitation efficacy assessment module calculates the patient's rehabilitation efficacy assessment coefficient based on electromyographic characteristic data; and judges the patient's recovery progress based on the rehabilitation efficacy assessment coefficient and the preset rehabilitation efficacy assessment threshold.

[0062] Obtain the patient's motion data, including:

[0063] The EEG cap collects real-time EEG signals from brain activity areas of patients during rehabilitation training using a rehabilitation robot.

[0064] Electromyography (EMG) signals of the affected limb are collected in real time by an EMG sensor attached to the skin surface of the muscles that move in the affected limb.

[0065] By performing time-domain and frequency-domain analyses on EEG and EMG signals, we obtained the EEG characteristic data corresponding to the EEG signals and the EMG characteristic data corresponding to the EMG signals.

[0066] Training artificial intelligence models based on historical motion data includes:

[0067] A1: Extract historical EEG and electromyography (EMG) feature data from historical motion data;

[0068] A2: By obtaining the mapping relationship between historical EEG feature data and motor intention data, as well as the mapping relationship between historical EMG feature data and motor intention data from the database; where motor intention data includes: forward, backward, sprinting and jogging, etc.

[0069] A3: An artificial intelligence model is trained based on historical EEG feature data to obtain a motor intention recognition model; an artificial intelligence model is trained based on historical EMG feature data to obtain a motor intention verification model; wherein, the artificial intelligence model includes: convolutional neural network or deep belief network;

[0070] A4: The motion recognition model is composed of a motion intent recognition model and a motion intent verification model.

[0071] An artificial intelligence model is trained based on historical EEG feature data to obtain a motor intention recognition model, including:

[0072] Historical EEG feature data is integrated into model input data, and the motion intention data corresponding to historical EEG feature data is integrated into model output data.

[0073] An artificial intelligence model is trained based on the model input data and model output data to obtain a motion intention recognition model.

[0074] An artificial intelligence model was trained based on historical electromyography (EMG) feature data to obtain a movement intention verification model, including:

[0075] Historical electromyography (EMG) feature data is integrated into standard input data, and the motion intention data corresponding to historical EMG feature data is integrated into standard output data.

[0076] An artificial intelligence model is trained based on standard input and standard output data to obtain a motion intent verification model.

[0077] Motion data is identified based on motion recognition models, including:

[0078] B1: Extract real-time acquired EEG and EMG feature data;

[0079] B2: Input the EEG feature data into the motor intention recognition model to obtain the recognition result;

[0080] B3: Input the electromyographic feature data into the motion intention verification model to obtain the verification and recognition results;

[0081] B4: Determine whether the verification result is consistent with the recognition result; if yes, output the recognition result; if no, return to step B1 until the verification result is consistent with the recognition result.

[0082] The rehabilitation efficacy evaluation coefficient for patients was calculated based on electromyographic data, including:

[0083] The duration of rehabilitation training for patients was obtained, and the root mean square value and zero crossover rate were extracted from the electromyography feature data.

[0084] The duration of rehabilitation training is denoted as T, the root mean square value as R, and the zero crossover rate as L.

[0085] Through the formula:

[0086] P=α / e^[(T-ZT)^2 / ZT]+β×e^[(R-CR) / ln(CR)]+ɡ×e^[(L-CL) / ln(CL)];

[0087] The rehabilitation efficacy evaluation coefficient is calculated; where P is the rehabilitation efficacy evaluation coefficient, α, β, and g are proportionality coefficients, and 0 < α < 1, 0 < β < 1, 0 < g < 1, ZT is the optimal rehabilitation training duration, CR is the initial root mean square value, and CL is the initial zero crossover rate.

[0088] The patient's recovery progress is assessed based on the rehabilitation efficacy evaluation coefficient and a preset rehabilitation efficacy evaluation threshold, including:

[0089] Determine whether the rehabilitation efficacy assessment coefficient is greater than or equal to the preset rehabilitation efficacy assessment threshold; if yes, it indicates that the patient's rehabilitation efficacy is good; if no, it indicates that the patient's rehabilitation efficacy is poor.

[0090] For example: Suppose there is a patient who needs to undergo rehabilitation training using a non-invasive brain-computer interface rehabilitation robot; specifically as follows:

[0091] Patient's basic information:

[0092] Initial condition: Lower limb paralysis due to stroke;

[0093] Initial root mean square value (CR): 50 μV;

[0094] Initial zero crossover rate (CL): 10 times / second;

[0095] The patient stands or sits on the exoskeleton robot, wearing an EEG cap and attaching electromyography (EMG) sensors. After the device is powered on, the display terminal shows prompts and illustrations, which the patient activates and deactivates by focusing their brain activity.

[0096] Data acquisition module;

[0097] 1. EEG signal acquisition:

[0098] The EEG cap was used to collect the patient's EEG signals in real time during rehabilitation training, with a focus on collecting EEG activity in the motor cortex.

[0099] The acquired EEG signals are analyzed in the time and frequency domains to obtain EEG characteristic data, including peak value, root mean square, zero crossover rate, signal amplitude, and signal power.

[0100] 2. Electromyography (EMG) signal acquisition:

[0101] Electromyography (EMG) sensors were attached to the quadriceps femoris and gastrocnemius muscles of the patient's lower limbs to collect EMG signals in real time.

[0102] The acquired electromyographic signals were analyzed in the time and frequency domains to obtain electromyographic characteristic data, including peak value, root mean square, zero crossover rate, signal amplitude, and signal power.

[0103] Data analysis module;

[0104] 1. Training the artificial intelligence model:

[0105] Historical data extraction: Extract historical EEG and EMG feature data, as well as corresponding movement intention data (such as forward, backward, sprint, and jog) from the database.

[0106] Model training:

[0107] Motor intention recognition model: A convolutional neural network (CNN) is trained using historical EEG feature data and corresponding motor intention data to obtain a motor intention recognition model.

[0108] Motor intention verification model: A deep belief network (DBN) is trained using historical electromyography feature data and corresponding motor intention data to obtain the motor intention verification model.

[0109] Combined model: Combines the motion intent recognition model and the motion intent verification model into a motion recognition model.

[0110] 2. Real-time data recognition:

[0111] Data extraction: Features are extracted from real-time acquired EEG and EMG feature data.

[0112] Model input:

[0113] The EEG feature data is input into the motion intention recognition model to obtain the recognition result.

[0114] The electromyographic feature data is input into the motion intent verification model to obtain the verification and recognition results.

[0115] Result Judgment: Determine whether the recognition result and the verification recognition result are consistent. If they are consistent, output the recognition result; if they are inconsistent, extract the data again for judgment until the recognition result and the verification recognition result are consistent.

[0116] Rehabilitation training module;

[0117] Training time: The patient underwent 45 minutes of rehabilitation training.

[0118] Recognition results: The motion intention recognition model identified the patient's motion intention as "moving forward", and the verification recognition model also confirmed this result.

[0119] The command to "move forward" is sent to the control terminal of the rehabilitation robot, which then controls the robot to perform "move forward" training.

[0120] Rehabilitation efficacy assessment module;

[0121] 1. Data extraction:

[0122] Rehabilitation training time (T): 45 minutes;

[0123] Root mean square value (R): 60 μV;

[0124] Zero crossover rate (L): 15 times / second;

[0125] 2. Calculate the rehabilitation efficacy evaluation coefficient:

[0126] Assuming the optimal rehabilitation training time (ZT) is 40 minutes;

[0127] Assume the proportionality constants are: α = 0.4, β = 0.3, g = 0.3;

[0128] Through the formula:

[0129] P=α / e^[(T-ZT)^2 / ZT]+β×e^[(R-CR) / ln(CR)]+ɡ×e^[(L-CL) / ln(CL)];

[0130] Calculate the rehabilitation efficacy evaluation coefficient;

[0131] P=0.4 / e^[(45-40)^2 / 40]+0.3×e^[(60-50) / ln(50)]+0.3×e^[(15-10) / ln(10)]≈6.7;

[0132] 3. Evaluation Results:

[0133] Assume the preset threshold for assessing rehabilitation efficacy is 6;

[0134] Since the rehabilitation efficacy evaluation coefficient (6.7) is greater than the preset rehabilitation efficacy evaluation threshold (6), it indicates that the patient's rehabilitation efficacy is good.

[0135] If the assessment result indicates that the patient's rehabilitation effect is poor, then appropriate measures should be taken to improve it, such as:

[0136] Adjust the intensity and duration of training, appropriately extend or shorten the training time, find the optimal rehabilitation training time, and at the same time adjust the intensity of training according to the patient's physical fitness and endurance.

[0137] Increase the diversity of training by introducing different types of training tasks, such as walking at different speeds and climbing stairs, to improve patients' interest and participation. Also, develop personalized training plans to ensure that the training content matches the patient's actual situation.

[0138] Provide psychological support and encouragement, help patients build confidence and overcome psychological barriers through psychological counseling, and establish a reward mechanism to encourage patients to actively participate in rehabilitation training and improve their enthusiasm and initiative in training;

[0139] These measures effectively improve the rehabilitation outcomes of patients and enhance the effectiveness of rehabilitation training.

[0140] The above examples demonstrate the significant benefits of this invention's non-invasive brain-computer interface rehabilitation robot in assessing and improving patient rehabilitation outcomes. Firstly, by collecting and analyzing patients' electroencephalogram (EEG) and electromyogram (EMG) signals in real time, the system can accurately identify patients' motor intentions and provide personalized rehabilitation training, thereby increasing patient engagement and training effectiveness. Secondly, by calculating rehabilitation efficacy evaluation coefficients, the system can scientifically assess patients' rehabilitation progress, promptly identify cases of poor rehabilitation outcomes, and take corresponding improvement measures, such as adjusting training intensity and duration, increasing training diversity, providing psychological support and motivation, strengthening rehabilitation assessment, and optimizing techniques. This comprehensive rehabilitation training method not only improves patient rehabilitation outcomes but also provides rehabilitation physicians and therapists with a scientific basis for assessment and adjustment, promoting the efficiency and personalization of rehabilitation training.

[0141] Some of the data in the above formula are calculated by removing dimensions and taking their numerical values. The formula is the closest to the real situation obtained by software simulation of a large amount of collected data. The preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

[0142] Working principle of the invention:

[0143] This invention addresses the technical problem that existing rehabilitation robots primarily use pre-set programs for repetitive rehabilitation training, where patients are not actively involved. This passive training method limits patients' autonomy and participation, resulting in limited rehabilitation efficacy, particularly in nerve damage repair. The invention involves acquiring patients' motion data; training an artificial intelligence model based on historical motion data to obtain a motion recognition model; identifying motion data using the motion recognition model to obtain recognition results; conducting rehabilitation training using a rehabilitation robot based on the recognition results; calculating the patient's rehabilitation efficacy evaluation coefficient based on the rehabilitation efficacy evaluation coefficient and a preset rehabilitation efficacy evaluation threshold; and determining the patient's recovery progress.

[0144] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A non-invasive brain-computer interface rehabilitation robot, characterized in that, The system includes a data acquisition module, a data analysis module, a rehabilitation training module, and a rehabilitation efficacy assessment module. Data acquisition module: used to acquire the patient's motion data; the motion data includes: electroencephalogram (EEG) characteristic data and electromyogram (EMG) characteristic data; Data analysis module: Trains an artificial intelligence model based on historical motion data to obtain a motion recognition model; uses the motion recognition model to identify motion data and obtain recognition results; Rehabilitation training module: Based on the recognition results, rehabilitation robots are used to provide rehabilitation training to patients; The rehabilitation efficacy assessment module calculates the patient's rehabilitation efficacy assessment coefficient based on electromyographic characteristic data; and judges the patient's recovery progress based on the rehabilitation efficacy assessment coefficient and the preset rehabilitation efficacy assessment threshold. The training of the artificial intelligence model based on historical motion data includes: A1: Extract historical EEG and electromyography (EMG) feature data from historical motion data; A2: By obtaining the mapping relationship between historical EEG feature data and motor intention data, as well as the mapping relationship between historical EMG feature data and motor intention data from the database; A3: An artificial intelligence model is trained based on historical EEG feature data to obtain a motor intention recognition model; an artificial intelligence model is trained based on historical EMG feature data to obtain a motor intention verification model; wherein, the artificial intelligence model includes: convolutional neural network or deep belief network; A4: The motion recognition model is composed of a motion intent recognition model and a motion intent verification model; The motion recognition model-based identification of motion data includes: B1: Extract real-time acquired EEG and EMG feature data; B2: Input the EEG feature data into the motor intention recognition model to obtain the recognition result; B3: Input the electromyographic feature data into the motion intention verification model to obtain the verification and recognition results; B4: Determine whether the verification result is consistent with the recognition result; if yes, output the recognition result; if no, return to step B1 until the verification result is consistent with the recognition result. The rehabilitation efficacy evaluation coefficient calculated based on electromyographic characteristic data includes: The duration of rehabilitation training for patients was obtained, and the root mean square value and zero crossover rate were extracted from the electromyography feature data. The duration of rehabilitation training is denoted as T, the root mean square value as R, and the zero crossover rate as L. Through the formula: P=α / e^[(T-ZT)^2 / ZT]+β×e^[(R-CR) / ln(CR)]+ɡ×e^[(L-CL) / ln(CL)]; The rehabilitation efficacy evaluation coefficient is calculated; where P is the rehabilitation efficacy evaluation coefficient, α, β, and g are proportionality coefficients, and 0 < α < 1, 0 < β < 1, 0 < g < 1, ZT is the optimal rehabilitation training duration, CR is the initial root mean square value, and CL is the initial zero crossover rate.

2. The non-invasive brain-computer interface rehabilitation robot according to claim 1, characterized in that, The acquisition of the patient's motion data includes: The EEG cap collects real-time EEG signals from brain activity areas of patients during rehabilitation training using a rehabilitation robot. Electromyography (EMG) signals of the affected limb are collected in real time by an EMG sensor attached to the skin surface of the muscles that move in the affected limb. By performing time-domain and frequency-domain analyses on EEG and EMG signals, we obtained the EEG characteristic data corresponding to the EEG signals and the EMG characteristic data corresponding to the EMG signals.

3. The non-invasive brain-computer interface rehabilitation robot according to claim 1, characterized in that, The artificial intelligence model trained based on historical EEG feature data to obtain a motion intention recognition model includes: Historical EEG feature data is integrated into model input data, and the motion intention data corresponding to historical EEG feature data is integrated into model output data. An artificial intelligence model is trained based on the model input data and model output data to obtain a motion intention recognition model.

4. The non-invasive brain-computer interface rehabilitation robot according to claim 1, characterized in that, The artificial intelligence model trained based on historical electromyography feature data, resulting in a motion intent verification model, includes: Historical electromyography (EMG) feature data is integrated into standard input data, and the motion intention data corresponding to historical EMG feature data is integrated into standard output data. An artificial intelligence model is trained based on standard input and standard output data to obtain a motion intent verification model.

5. The non-invasive brain-computer interface rehabilitation robot according to claim 1, characterized in that, The method of judging the patient's recovery progress based on the rehabilitation efficacy assessment coefficient and the preset rehabilitation efficacy assessment threshold includes: Determine whether the rehabilitation efficacy assessment coefficient is greater than or equal to the preset rehabilitation efficacy assessment threshold; if yes, it indicates that the patient's rehabilitation efficacy is good; if no, it indicates that the patient's rehabilitation efficacy is poor.

Images