A personalized feedback adjustment method for motor imagery brain-computer interface

Abstract

The application provides a personalized brain-controlled mechanical arm rehabilitation system based on individual characteristics, comprising the following steps: S1, evaluating the performance of each MI-BCI training cycle of an individual; S2, researching the correlation between the individual's electroencephalogram features and functional indicators; S3, predicting the motor imagery ability according to the cycle rating and electroencephalogram feature quantitative data of the individual, and then establishing a mapping from the motor imagery ability to the mechanical arm movement rate; S4, constructing a personalized feedback regulation loop and establishing a complete closed-loop MI-BCI mechanical arm system. The application has the beneficial effects that: the personalized brain-controlled mechanical arm system based on individual characteristics predicts the motor imagery ability according to the performance of the training cycle and the electroencephalogram features, thereby predicting the mechanical arm rate, and the MI-BCI ability of the individual is displayed in real time by the mechanical arm rate. The individual's initiative is increased, the degree of assistance of the MI-BCI to the individual is reduced, and the rehabilitation effect is improved.

Description

Technical Field

[0001] This invention belongs to the field of brain-computer interface technology in biomedical engineering, and in particular relates to a personalized feedback regulation method for brain-computer interfaces for motor imagination. Background Technology

[0002] Brain-computer interfaces (BCIs) are direct information pathways that transmit information by connecting the central nervous system and a computer, without relying on the peripheral nervous system. Motor imagery (MI), as one of the main BCI paradigms, is widely used in the field of motor rehabilitation. It interprets brain signals to obtain motor intentions, which helps to repair motor neural circuits. Using BCI rehabilitation robot systems, an individual's intentions can be translated into rehabilitation assistive commands or interventions, thereby reconstructing the individual's interaction with the outside world and the rehabilitation circuit.

[0003] Several issues remain to be addressed in MI-BCI rehabilitation robot technology. For example, individual differences exist in brain electrical activity and the rehabilitation process, but traditional BCI rehabilitation systems employ fixed training programs, lacking flexible feedback mechanisms and failing to promptly reflect individual characteristics and training needs. Furthermore, the lack of personalized closed-loop feedback adjustment may lead to poor training outcomes or excessive fatigue. These problems limit the application and promotion of BCI rehabilitation technology.

[0004] This invention proposes a personalized feedback adjustment method for the MI-BCI robotic arm. It establishes dynamic feedback on the robotic arm's rate based on individual training cycle ratings and EEG characteristics, and adjusts the degree of assistance the MI-BCI system provides to the brain in controlling the robotic arm's movement, thus creating a personalized closed-loop system. This method can effectively adapt to the characteristics of brain cognition and BCI training patterns to construct personalized robotic arm feedback, solving the personalization problem of MI-BCI and promoting its practical application in rehabilitation. Summary of the Invention

[0005] In view of this, the present invention aims to propose a personalized feedback adjustment method for a MI-BCI robotic arm system based on individual characteristics. By studying the periodic characteristics and EEG features of individuals, a quantitative basis for individual characteristics based on training cycle rating and real-time EEG features is proposed; based on this, the motion rate of the robotic arm is predicted to achieve real-time output of dynamic feedback; after the periodic training ends, the cycle level needs to be updated to achieve cyclical training and evaluation, establishing a closed-loop MI-BCI system and steadily improving the training effect of MI-BCI.

[0006] To achieve the above objectives, the technical solution of the present invention is implemented as follows:

[0007] A personalized feedback adjustment method for an MI-BCI robotic arm system includes the following steps:

[0008] S1. Quantitatively evaluate the performance of individuals in each MI-BCI training cycle;

[0009] S2. Further research on real-time assessment methods for individual EEG characteristics based on periodic rating;

[0010] S3. Establish a robotic arm motion rate mapping function based on individual periodic ratings and EEG characteristic data to achieve personalized robotic arm motion feedback.

[0011] Furthermore, in step S1, MI controllability can be used as a quantitative indicator for the rating of the motion imagery cycle.

[0012] Furthermore, MI controllability refers to an individual's ability to generate clear, accurate, and stable internal representations during motor imagery. The motor imagery controllability test was used as a subjective indicator to quantify MI controllability; the classification accuracy of the MI EEG decoding model was used as an objective indicator to quantify MI controllability.

[0013] Furthermore, in step S2, the quantitative indicators are first unified, and then the signal based on MI frequency domain energy is evaluated.

[0014] Furthermore, the process of assessing EEG characteristics includes the following steps:

[0015] S201. Collect multi-channel brain data, perform spatial filtering on 2 to 6 seconds of EEG data during the subject's MI, and calculate spectral energy characteristics.

[0016] S202. Based on the changes in spectral energy in the resting state and MI mission state, make personalized selections of channels and frequencies;

[0017] S203. Select the amplitude of the channels and frequencies most related to MI on the left and right sides of the brain, respectively, as signal features, and convert the EEG signal features into evaluation values.

[0018] Furthermore, in step S3, a predictor for the transformation from cycle level and EEG characteristics to robotic arm movement rate is constructed based on the MI training cycle rating and EEG signal intensity quantification.

[0019] Furthermore, completing the mapping from periodic levels and EEG characteristics to robotic arm movement rates includes the following steps:

[0020] S301. Set a threshold to convert real-time EEG energy into an evaluation value;

[0021] S302. Map the periodic rating results and real-time EEG features to the control rate of the robotic arm.

[0022] S303. Update the weight ratio of EEG feature intensity and periodic rating according to the periodic level to achieve feedback intensity update: gradually reduce the weight of periodic rating, and transfer the control of the robotic arm to human active imagination in stages to adapt to the enhancement of human MI ability; at the same time, gradually increase the weight of EEG feature intensity to strengthen the perception and control of one's own imagination activities.

[0023] Compared with existing technologies, the personalized feedback adjustment method of the MI-BCI robotic arm system based on individual characteristics described in this invention has the following advantages:

[0024] (1) The method described in this invention is a method for quantitatively evaluating the performance of an individual in each MI-BCI training cycle. By integrating subjective and objective quantitative indicators that can characterize the controllability of motor imagery, a rating method for MI-BCI training cycles is established, providing a quantitative basis for personalized feedback from the perspective of motor imagery training cycles, and re-evaluating and updating the training cycle level of MI-BCI after the cycle training ends, thereby realizing cyclical training and rating;

[0025] (2) The method described in this invention further studies the real-time quantification method of individual EEG characteristics based on periodic rating. It analyzes EEG characteristics that can characterize the intensity of motor imagery, extracts these characteristics to construct a quantitative index based on EEG intensity, and provides a quantitative basis for personalized feedback from the perspective of real-time EEG characteristics;

[0026] (3) The method described in this invention is a method for establishing a mapping of robotic arm movement rate based on individual period rating and EEG characteristic quantitative data. The mapping relationship between period level and EEG intensity and rate is established respectively, and the weight law of the influence of the above two on the robotic arm rate during long-term MI-BCI training is studied. Finally, a complete dynamic mapping is established to realize personalized feedback adjustment of robotic arm movement rate. Attached Figure Description

[0027] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0028] Figure 1 This is an example diagram of the imagined control item of the CMI test, a subjective evaluation index for periodic rating, as described in an embodiment of the present invention.

[0029] Figure 2 This is the electroencephalogram (EEG) feature assessment diagram described in the embodiments of the present invention;

[0030] Figure 3 This is a flowchart of the dynamic feedback process of the robotic arm based on individual training characteristics as described in an embodiment of the present invention;

[0031] Figure 4 The electrode placement position of the 32 conductive electrode cap in the method described in this embodiment of the invention;

[0032] Figure 5 shows the experimental environment of the method described in the embodiment of the present invention. Detailed Implementation

[0033] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0034] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0035] like Figure 1 As shown in Figure 5, a personalized feedback adjustment method for an MI-BCI robotic arm system based on individual characteristics includes the following steps:

[0036] S1. Construct a rating method for the training cycle of motor imagery;

[0037] S2. Constructing an individual EEG characteristic assessment method;

[0038] S3. Construct a personalized feedback adjustment method based on robotic arm motion mapping.

[0039] In step S1, the motion imagery cycle rating is evaluated by integrating subjective and objective quantitative indicators of MI controllability.

[0040] The subjective metric to be used is the Controllability of Motor Imagery Test (CMI Test). The CMI Test consists of M task items, each representing a sequence of actions containing N imaginary controls (i.e., actions performed sequentially). The action cues corresponding to the controls are defined as test instructions. The evaluation steps are as follows:

[0041] S301. Subjects are asked to first imagine the first imaginary control item in the task;

[0042] S302. Next, imagine the subsequent control items according to the test instructions.

[0043] S303. After completing the visualization of the control item, perform a matching test: Select the option from five options (including four action sequences and a forget option) that has the same meaning as the action in the test instructions.

[0044] S104. The controllability of the subject's MI is assessed based on the matching accuracy of completing M task items. The higher the matching accuracy, the stronger the individual's controllability of their self-body imagery ability.

[0045] The objective indicator to be adopted is the classification accuracy of the decoding model of MI EEG as an objective quantitative indicator of MI controllability. The classification accuracy refers to the ratio between the number of correctly classified MI trials and the total number of trials. The higher the classification accuracy value, the stronger the controllability of the individual to generate identifiable EEG features.

[0046] The specific implementation is as follows:

[0047] (1) Dynamic feedback based on individual characteristics

[0048] Based on the MI training cycle rating and EEG feature assessment, a mapping from cycle level and EEG signal intensity to robotic arm movement rate is further constructed to achieve dynamic feedback based on individual characteristics.

[0049] Feedback refers to the results related to one's mental activity (MI) presented to a person through the BCI system. Motion feedback mapping maps individual characteristics to the movements of a robotic arm, using speed feedback to inform the person's imaginative abilities. The mapping function updates the weighting of EEG intensity and periodic rating based on periodic levels, thus updating the feedback intensity: gradually decreasing the weight of periodic rating to gradually transfer control of the robotic arm to the person's active imagination, adapting to the enhancement of MI abilities; simultaneously, gradually increasing the weight of EEG intensity to strengthen the perception and control of one's own imaginative activities.

[0050] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

[0051] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A personalized feedback adjustment method for motor imagery brain-computer interface, characterized in that, The method comprises the following steps: S1, quantitatively evaluating the performance of each motor imagery brain-computer interface training cycle of an individual to obtain a training cycle rating result; S2, evaluating the characteristics of the electroencephalogram signals generated during the motor imagery process to obtain electroencephalogram characteristic data for representing the intensity of motor imagery; S3, on the basis of the training cycle rating result and the electroencephalogram characteristic data, establishing a mapping relationship between the movement rate of the mechanical arm and the training cycle rating result and the electroencephalogram characteristic data, and outputting the corresponding movement rate of the mechanical arm according to the mapping relationship, wherein the training cycle rating result is used to represent the performance of the individual at the training cycle level, and the electroencephalogram characteristic data is used to represent the change in electroencephalogram characteristics during the motor imagery process; S4, during the training process, re-evaluating the training cycle rating result according to the update of the training cycle, and gradually reducing the influence weight of the training cycle rating result on the movement rate of the mechanical arm and gradually increasing the influence weight of the electroencephalogram characteristic data on the movement rate of the mechanical arm during the adjustment process of the movement rate of the mechanical arm, so as to realize the update of the feedback intensity and the phased transfer of the control dominance of the mechanical arm from the cycle rating to the electroencephalogram characteristic, thereby constructing a personalized feedback adjustment loop for the motor imagery brain-computer interface.

2. The personalized feedback adjustment method for the motor imagery brain-computer interface according to claim 1, characterized in that the quantitative evaluation of the training cycle rating comprises subjective indicators and objective indicators, wherein the subjective indicators are motor imagery controllability test results, and the objective indicators are classification accuracies based on a motor imagery electroencephalogram signal decoding model.

3. The personalized feedback adjustment method for the motor imagery brain-computer interface according to claim 1, characterized in that the electroencephalogram characteristic data is electroencephalogram characteristics based on motor imagery frequency energy.

4. The personalized feedback adjustment method for the motor imagery brain-computer interface according to claim 3, characterized in that the electroencephalogram characteristic evaluation comprises the following steps: S201, collecting multi-channel electroencephalogram signals, spatially filtering the electroencephalogram data during the motor imagery stage, and calculating the frequency spectrum energy characteristics; S202, individually selecting the channels and frequencies of the electroencephalogram signals according to the changes in the frequency spectrum energy in the resting state and the motor imagery task state; S203, selecting the amplitudes of the left and right brain channels and the corresponding frequencies related to motor imagery as the electroencephalogram characteristics, and converting the electroencephalogram characteristics into numerical values for evaluation.

5. The personalized feedback adjustment method for the motor imagery brain-computer interface according to claim 1, characterized in that the electroencephalogram characteristic data is used to represent the motor imagery ability of the individual and serves as the basis for adjusting the movement rate of the mechanical arm.

6. The personalized feedback adjustment method for the motor imagery brain-computer interface according to claim 1, characterized in that the mapping process of the movement rate of the mechanical arm comprises: mapping the training cycle rating result and the electroencephalogram characteristic data to the control rate of the mechanical arm. ​ ​ ​ ​ ​ During the training process, the weighting of training cycle rating results and EEG feature data in rate regulation was gradually adjusted, so that the influence of training cycle rating results gradually decreased and the influence of EEG feature data gradually increased.

Images