Motor imagery classification method based on electroencephalogram traceability and dipole selection

Abstract

The invention provides a motor imagery classification method based on electroencephalogram tracing and dipole selection. The motor imagery classification method comprises the steps that electroencephalogram signals based on multi-class motor imagery are collected; tracing the multichannel electroencephalogram signal to obtain a signal of a cortical neural activity source; dipole channel selection is carried out on source space dipoles, energy of electroencephalogram signals of all dipoles is used as a search strategy for selecting and deleting a dipole channel set, and improved F-score values of electroencephalogram signal energy of all motor imagery categories and remaining categories are combined to be used as an optimal dipole channel selection evaluation criterion; extracting electroencephalogram data of the source space selection dipole; inputting the electroencephalogram data into a common spatial mode filter for feature extraction; and inputting the co-spatial pattern features into a support vector machine classifier to realize motor imagery electroencephalogram signal classification. On the basis of exploring the motor imagery electroencephalogram law, motor imagery electroencephalogram signal processing, feature extraction and classification method research is carried out, and the classification accuracy is effectively improved.

Description

technical field [0001] The invention relates to the technical field of EEG signal processing, in particular to a motor imagery classification method based on EEG traceability and dipole selection, which is used to improve the accuracy of motor imagery EEG signal task decoding. Background technique [0002] Brain-computer interface refers to the direct connection created between the human or animal brain and external devices to realize the exchange of information between the brain and the device. By analyzing the EEG signal, detecting and identifying the activation effect of different brain regions to judge the user's intention, understand the information processing process of the human brain, and then realize the communication and control between the brain and external devices. [0003] Brain-computer interface based on motor imagery is an important part of many brain-computer interface paradigms. This paradigm needs to collect EEG signals from subjects when they perform spe...

AI Technical Summary

Problems solved by technology

The first category is the introduction of EEG technology, mainly fMRI technology, but the brain-computer interface based on fMRI is not suitable for daily use due to the large time delay

The second type is invasive EEG signal acquisition, but this method is likely to trigger the patient's immune response and callus, which will lead to the decline of signal quality until it disappears

[0005] In the prior art, the correct rate of multiple classifications of motor imagery needs to be improved urgently. The analysis process of motor imagery EEG signals includes: signal acquisition, preprocessing, feature extraction and selection, and pattern recognition, where the amplitude of EEG signals detected by scalp electrodes is weak , is susceptible to noise interference, and the spatial resolution of EEG is poor, and the difference between subjects is obvious. Therefore, the research on multi-category classification of motor imagery needs to be further developed

Images