Motor imagery training method and system based on combination of pain feedback and storage medium

Abstract

The invention discloses a motor imagery training method and system based on combination of pain feedback and a storage medium, and the method comprises the steps: guiding a user to execute a motor imagery task, and stimulating the limb of the affected side of the user through a stimulation source; collecting electroencephalogram data in the motor imagery task executed by the user; inputting the electroencephalogram data into a pain feature model obtained by pre-training to obtain a pain recognition result; according to the pain recognition result, if it is judged that the user is painful and the output value of the current stimulation source is larger than a stimulation source threshold value, the maximum value of the output value of the stimulation source is limited to be the stimulation source threshold value; within the set time after the motor imagery task is finished, myoelectricity data of the limb on the affected side are collected, and whether the stimulation mode of the stimulation source is changed or not is judged according to the myoelectricity data. According to the method and the device, the optimal stimulation intensity which can stimulate the motor imagery and cannot cause excessive pain can be found, so that the task setting and stimulation conditions of the motor imagery are optimized, and the normal form and strategy of the motor imagery are further optimized.

Description

Technical Field

[0001] The present application relates to the field of electroencephalogram (EEG) data processing, and in particular to a motor imagery training method and system based on combining pain feedback. Background Art

[0002] The application of brain-computer interfaces in medicine has become increasingly extensive with the advancement of social science and technology, including diagnosing diseases through EEG data, or using EEG paradigms to help patients undergo rehabilitation treatment for certain specific conditions.

[0003] Motor imagery is a common EEG application paradigm that mainly helps patients to actively imagine to achieve neurorehabilitation treatment. However, the current motor imagery paradigm has strong requirements for environmental influences and paradigm presentation methods in clinical trials. If the environment interferes or the paradigm procedure is not good enough, the effect of the experiment will be greatly reduced. Summary of the invention

[0004] The technical purpose of the present application is to provide a motor imagery training method based on combining pain feedback, so as to optimize the motor imagery paradigm and improve the training effect.

[0005] In order to achieve the above technical objectives, this application adopts the following technical solutions.

[0006] In a first aspect, the present application provides a motor imagery training method based on combining pain feedback, comprising: guiding the user to perform a motor imagery task, while stimulating the affected limb of the user using a stimulation source; Collecting EEG data of the user performing the motor imagery task; Inputting the EEG data into a pre-trained pain feature model to obtain a pain recognition result; If, according to the pain recognition result, it is determined that the user is in pain and the current output value of the stimulus source is greater than a stimulus source threshold, the maximum value of the output value of the stimulus source is limited to the stimulus source threshold; Within a set time after the motor imagery task is completed, the electromyographic data of the affected limb is collected to determine whether to change the stimulation mode of the stimulation source according to the electromyographic data.

[0007] In some embodiments, the method further comprises pre-training to obtain the pain feature model, specifically comprising: Using a stimulation source to stimulate the affected limb of the user while the user performs a motor imagery task; Collecting EEG data of the user when there is no pain and when there is pain when the user performs the motor imagery task; Using a common space model to extract features from all the EEG data; The extracted features and corresponding category labels are input into a classification model for training to obtain the pain feature model.

[0008] In some embodiments, determining whether to change the stimulation mode of the stimulation source according to the electromyographic data includes: Determine the time difference between the maximum value and the minimum value of the electromyographic data; When the time difference is greater than the muscle recovery time threshold, the stimulation mode of the stimulation source is changed.

[0009] In some embodiments, the motor imagery training method further includes determining the stimulus source threshold, specifically including: The stimulation source is used to stimulate the affected limb of the user, and when the user feels pain, the output value corresponding to the stimulation source is determined as the upper limit of the output value of the stimulation source; The output value upper limit is decreased by a set amount and is determined as the stimulus source threshold.

[0010] In some embodiments, the motor imagery training method also includes updating the stimulus source threshold, specifically including: based on the pain recognition result, if it is determined that the user is not in pain, and the current output value of the stimulus source is greater than the output value upper limit, then the current output value of the stimulus source is used as the latest output value upper limit; the latest output value upper limit is reduced by a set amplitude and determined as the latest stimulus source threshold.

[0011] In some embodiments, stimulating the affected limb of the user using a stimulation source includes: using the changes in the amplitude and frequency of the output value of the stimulation source to cause the affected limb of the patient to perform the same guided movement changes as the motor imagery task.

[0012] In a second aspect, the present application provides a motor imagery application device based on pain feedback, comprising: A motor imagery task guidance module is used to guide the user to perform a motor imagery task; A muscle stimulation module, used for stimulating the affected limb of the user while the motor imagery task guidance module guides the user to perform the motor imagery task; An EEG data collection module is used to collect EEG data of the user performing the motor imagery task; A pain recognition module, used for inputting the EEG data into a pre-trained pain feature model to obtain a pain recognition result; A stimulation source adjustment module, configured to limit the maximum value of the output value of the stimulation source of the muscle stimulation module to the stimulation source threshold value if it is determined that the user is in pain and the current output value of the stimulation source is greater than the stimulation source threshold value according to the pain recognition result; The electromyographic data recognition module is used to collect the electromyographic data of the affected limb within a set time after the motor imagery task is completed, so as to determine whether to change the stimulation mode of the stimulation source according to the electromyographic data.

[0013] In a third aspect, an embodiment of the present application further provides a computer-readable storage medium storing a computer program, which, when executed, can implement a motor imagery training method based on combined pain feedback as provided in any possible implementation manner of the first aspect.

[0014] Compared with the prior art, the motor imagery training method based on pain feedback provided by the embodiment of the present application has the following beneficial technical effects: by real-time monitoring of the user's pain feedback and dynamically adjusting the output value of the stimulus source, this immediate feedback mechanism helps to find the optimal stimulation intensity that can stimulate motor imagery without causing excessive pain, thereby optimizing the task setting and stimulation conditions of motor imagery; the timely processing of pain feedback reduces the user's discomfort during training and increases their comfort and satisfaction. In addition, everyone's pain threshold and reaction are different. This method provides a customized training program for each user through personalized pain identification and adjustment. In addition to pain feedback, the method also combines the collection and analysis of electromyographic data to evaluate the training effect and adjust the stimulation method in time. This comprehensive evaluation and adjustment mechanism helps to more comprehensively understand the user's training status and effect, thereby further optimizing the paradigm and strategy of motor imagery.

[0015] Compared with the prior art, the motor imagery training system based on pain feedback provided in the embodiment of the present application has the same beneficial technical effects as the above-mentioned motor imagery training method based on pain feedback, which will not be elaborated. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are for explanation purposes only and are not intended to limit the scope of the present application in any way. In addition, the shapes and proportional dimensions of the components in the drawings are only for illustration purposes and are used to help understand the present application. They do not specifically limit the shapes and proportional dimensions of the components of the present application. Under the guidance of the present application, those skilled in the art can select various possible shapes and proportional dimensions to implement the present application according to specific circumstances. In the drawings: Figure 1 Schematic diagram of a flow chart of a motor imagery training method based on combining pain feedback in an embodiment; Figure 2Schematic diagram of the training process of the pain feature model in the embodiment; Figure 3 A schematic diagram of an electroencephalogram data waveform under stimulation by a stimulus source in one embodiment; Figure 4 is a schematic diagram of an electroencephalogram data waveform under stimulation by a stimulus source in another embodiment;

[0017] Figure 5 Schematic diagram of the system structure of the motor imagery training method based on pain feedback in the embodiment. DETAILED DESCRIPTION

[0018] In order to enable those skilled in the art to better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this application.

[0019] The present application provides a motor imagery training method based on pain feedback, which improves the rehabilitation effect of motor imagery by optimizing the paradigm.

[0020] The present application is further described below in conjunction with the accompanying drawings and specific embodiments.

[0021] like Figure 1 As shown, a motor imagery training method based on combined pain feedback includes: guiding a user (patient) to perform a motor imagery task, while stimulating the affected limb of the user with a stimulation source; Collect EEG data of users performing motor imagery tasks; Input the EEG data into the pre-trained pain feature model to obtain the pain recognition result; According to the pain recognition result, if it is determined that the user is in pain and the output value of the current stimulus source is greater than the stimulus source threshold, the maximum value of the output value of the stimulus source is limited to the stimulus source threshold; Within a set time after the motor imagery task is completed, the electromyographic data of the affected limb is collected to determine whether to change the stimulation method of the stimulation source based on the electromyographic data.

[0022] In a specific implementation, the user wears an electroencephalogram (EEG) acquisition device and an electromyography (EMG) acquisition device, and wears an auxiliary stimulation source, which may be electrical stimulation or mechanical stimulation.

[0023] Training is started through the motor imagery paradigm. EEG data needs to be collected during the imagination guidance stage, and physical guidance is performed during the collection process based on the auxiliary stimulation source. In addition, pain monitoring is required during the auxiliary stimulation process to adjust the intensity of the stimulation source in real time. During the rest process after the imagination ends, electromyographic data is collected and analyzed to see whether the stimulation method of the imagination process is adapted by the muscles, thereby changing the frequency and method of stimulation.

[0024] This application uses EEG acquisition equipment, combined with the motor imagery paradigm, guided image content, and auxiliary stimulation to help patients conduct more effective training and ensure the comprehensiveness of the training; through electromyography monitoring, the degree of muscle response to stimulation can be monitored in real time, and timely and accurate feedback adjustments can be made; by monitoring pain responses, the stimulation intensity can be corrected in time to avoid causing discomfort to the patient and to achieve treatment more efficiently.

[0025] Some embodiments also include pre-training to obtain a pain feature model, such as Figure 2 As shown, it specifically includes: in the offline stage, while the user performs a motor imagery task (such as motor imagery of the left and right hands), the stimulation source is used to stimulate the user's affected limb; the EEG data when the user performs the motor imagery task without pain and the EEG data when the user is in pain are collected; the common space mode is used to extract features from all the EEG data; the extracted features are input into the support vector machine for training to obtain a pain feature model. In the online stage, different paradigm models can be compared, for example, using a data model without an auxiliary stimulation source and a data model with an auxiliary stimulation source to compare the classification accuracy, so as to change the training mode.

[0026] In the specific implementation, the motor imagery task is combined with auxiliary stimulation for guidance. The single motor imagery paradigm includes an imagination phase and a rest phase.

[0027] In some embodiments, stimulating the affected limb of the user using a stimulation source includes: using the changes in the amplitude and frequency of the output value of the stimulation source to make the affected limb of the patient perform the same guided movement changes as in the motor imagery task.

[0028] In the imagination stage of the motor imagery paradigm, the patient can enter the motor imagery stage through animation information prompts and voice prompts. In addition, auxiliary stimulation sources can be used to perform corresponding physical activities of the affected limb, such as "clenching a fist" and "opening" to achieve cognitive learning in a more realistic prompting manner.

[0029] Three steps need to be completed during motor imagery. The first is to synchronize the visual guidance with the frequency and method of auxiliary stimulation. For example, during the 4-second imagination process, the stimulation method used is as follows: Figure 3If the image is "clench fist-open-clench fist-open", the animation prompt also needs to prepare animation effects of corresponding frequencies. If the stimulation method during the 4-second imagination process is as follows Figure 4 If the "fist" is shown, the animation also needs to be presented at the same frequency.

[0030] However, the guidance of the auxiliary stimulation source is mainly physical stimulation, which will cause the affected limb to move. If the degree of stimulation is uncomfortable during the exercise, it will affect the patient's experience and the quality of the data.

[0031] In some embodiments, the threshold of the stimulus source needs to be determined in advance. The specific operation is to stimulate the affected limb of the patient through the stimulus source in the early stage of training, and determine the output value of the stimulus source when the patient feels pain, then determine the output value corresponding to the stimulus source as the upper limit of the output value of the stimulus source; and determine the stimulus source threshold after the upper limit of the output value decreases by a set amplitude.

[0032] Based on electromyographic judgment, the pain threshold can be broken through through real-time EEG monitoring, thereby achieving faster recovery effects.

[0033] In some embodiments, the motor imagery training method also includes updating the stimulus source threshold, specifically including: based on the pain recognition result, if it is determined that the user is not in pain, and the output value of the current stimulus source is greater than the output value upper limit, then the output value of the current stimulus source is used as the latest output value upper limit; the latest output value upper limit is reduced by a set amplitude and determined as the latest stimulus source threshold.

[0034] This application uses electromyography or electroencephalography to perform real-time pain monitoring, and uses this method to set the threshold of the stimulus source intensity, which can reduce the user's discomfort during training and increase their comfort and satisfaction. In addition, everyone's pain threshold and reaction are different. This method provides a customized training plan for each user through personalized pain identification and adjustment.

[0035] Since auxiliary stimulation is performed during the imagination process, a certain contraction may be produced to the muscles at this time. Because the sensitivity of muscles to stimulation will gradually weaken with the growth of time, resulting in a lower and lower degree of contraction, this activity can be determined by monitoring the degree of contraction of muscles through electromyography. In the embodiment, when the electromyographic data within the preset time after the end of the motor imagination stage is collected, the stimulation source also stops stimulating, so the electromyographic data of the corresponding part can be collected within the preset time when the resting begins. According to the time difference between the maximum value and the minimum value of the electromyographic data, it can be judged whether the user has adapted to the stimulation, so as to determine whether to change the stimulation mode of the stimulation source. When the muscles are first electrically stimulated, since the muscle cells are more sensitive to the reaction of electrical stimulation, they will quickly change from a resting state to an excited state and generate a strong electromyographic signal. However, with the increase in the number of electrical stimulations and the gradual adaptation of the intensity, the sensitivity of muscle cells to electrical stimulation will gradually decrease, and the speed at which muscle cells recover from an excited state to a resting state will slow down, resulting in an increase in the time required for the electromyographic signal to change from a maximum value to a minimum value, and then the stimulation mode of the stimulation source will be changed.

[0036] The embodiment of the present application performs electromyography monitoring on the rehabilitation part through electromyography acquisition equipment, and determines whether the current rehabilitation part is adaptable to the stimulation source, which helps to more comprehensively understand the user's training status and effect, thereby further optimizing the paradigm and strategy of motor imagery.

[0037] Based on the same inventive concept as the above-mentioned motor imagery training based on combining pain feedback, the present application embodiment also provides a motor imagery training system based on combining pain feedback, such as Figure 5 As shown, it includes: a motor imagery task guiding module, a muscle stimulation module, an EEG data acquisition module, a pain recognition module, a stimulation source adjustment module and an EMG data recognition module.

[0038] Among them, the motor imagery task guidance module is used to guide the user to perform the motor imagery task.

[0039] The muscle stimulation module is used to stimulate the user's affected limb while the motor imagery task guidance module guides the user to perform the motor imagery task.

[0040] The EEG data acquisition module is used to collect EEG data of users performing motor imagery tasks.

[0041] The pain recognition module is used to input EEG data into a pre-trained pain feature model to obtain pain recognition results.

[0042] The stimulation source adjustment module is used to limit the maximum value of the output value of the muscle stimulation module stimulation source to the stimulation source threshold if it is judged that the user is in pain and the output value of the current stimulation source is greater than the stimulation source threshold based on the pain recognition result.

[0043] The electromyographic data recognition module is used to collect the electromyographic data of the affected limb within a set time after the motor imagery task is completed, so as to determine whether to change the stimulation mode of the stimulation source according to the electromyographic data.

[0044] The motor imagery training system based on pain feedback provided by the embodiment of the present application analyzes the user's EEG data in real time through the pain recognition module, and dynamically adjusts the output value of the muscle stimulation module according to the user's actual pain perception. This personalized adjustment mechanism ensures the safety and effectiveness of the training process and avoids the risk of aggravating the user's pain due to excessive stimulation. The system realizes real-time monitoring of the user's pain perception through EEG data collection and pain recognition technology, and can make dynamic adjustments based on the user's actual feelings, thereby improving the patient's participation and comfort during the training process.

[0045] The embodiment of the present application also provides a computer-readable storage medium storing a computer program. When the computer program is executed, the above-mentioned motor imagery training method based on combining pain feedback can be implemented.

[0046] It should be noted that in the present application, the emphasis of the schemes described in various embodiments is different, but the various embodiments are also somewhat interrelated. When understanding the scheme of the present application, the various embodiments can refer to each other.

[0047] The above is a detailed introduction to the motor imagery training method, system and storage medium based on pain feedback provided by the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the concept of the present application and should not be understood as limiting the scope of protection of the present application.

Claims

1. A motor imagery training method based on pain feedback, characterized in that: include: guiding the user to perform a motor imagery task, while stimulating the affected limb of the user using a stimulation source; Collecting EEG data of the user performing the motor imagery task; Inputting the EEG data into a pre-trained pain feature model to obtain a pain recognition result; If, according to the pain recognition result, it is determined that the user is in pain and the current output value of the stimulus source is greater than a stimulus source threshold, the maximum value of the output value of the stimulus source is limited to the stimulus source threshold; Within a set time after the motor imagery task is completed, the electromyographic data of the affected limb is collected to determine whether to change the stimulation mode of the stimulation source according to the electromyographic data.

2. The motor imagery training method according to claim 1, characterized in that: The method further includes pre-training to obtain the pain feature model, specifically including: Using a stimulation source to stimulate the affected limb of the user while the user performs a motor imagery task; Collecting EEG data of the user when there is no pain and when there is pain when the user performs the motor imagery task; Using a common space model to extract features from all the EEG data; The extracted features and corresponding category labels are input into a classification model for training to obtain the pain feature model.

3. The motor imagery training method according to claim 1, characterized in that: Determining whether to change the stimulation mode of the stimulation source according to the electromyographic data includes: Determine the time difference between the maximum value and the minimum value of the electromyographic data; When the time difference is greater than the muscle recovery time threshold, the stimulation mode of the stimulation source is changed.

4. The motor imagery training method according to claim 1, characterized in that: The motor imagery training method further includes determining the stimulus source threshold, specifically including: The stimulation source is used to stimulate the affected limb of the user, and when the user feels pain, the output value corresponding to the stimulation source is determined as the upper limit of the output value of the stimulation source; The output value upper limit is decreased by a set amount and is determined as the stimulus source threshold.

5. The motor imagery training method according to claim 4, characterized in that: The motor imagery training method further includes updating the stimulus source threshold, specifically including: If, according to the pain recognition result, it is determined that the user has no pain, and the current output value of the stimulation source is greater than the output value upper limit, the current output value of the stimulation source is used as the latest output value upper limit; The latest upper limit of the output value is reduced by a set amount and is determined as the latest stimulus source threshold.

6. The motor imagery training method according to claim 1, characterized in that: Stimulating the affected limb of the user by using a stimulation source, including: The changes in the amplitude and frequency of the output value of the stimulation source are used to cause the affected limb of the patient to perform the same guided movement changes as in the motor imagery task.

7. A motor imagery training system based on pain feedback, characterized in that: include: A motor imagery task guidance module is used to guide the user to perform a motor imagery task; A muscle stimulation module, used for stimulating the affected limb of the user while the motor imagery task guidance module guides the user to perform the motor imagery task; An EEG data collection module is used to collect EEG data of the user performing the motor imagery task; A pain recognition module, used for inputting the EEG data into a pre-trained pain feature model to obtain a pain recognition result; A stimulation source adjustment module, configured to limit the maximum value of the output value of the stimulation source of the muscle stimulation module to the stimulation source threshold value if it is determined that the user is in pain and the current output value of the stimulation source is greater than the stimulation source threshold value based on the pain recognition result; The stimulation source adjustment module is used to collect the electromyographic data of the affected limb within a set time after the motor imagery task is completed, so as to determine whether to change the stimulation mode of the stimulation source according to the electromyographic data.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed, the motor imagery training method based on combining pain feedback as described in any one of claims 1 to 6 can be implemented.

Images