Binary classification motor imagery EEG recognition method based on interpretable clustering model

Abstract

The present invention provides a binary motor imagery EEG signal recognition method based on an interpretable clustering model, which includes step 1, obtaining multi-channel motor imagery EEG data of a subject, successively truncating and filtering the EEG data, and storing them as High-dimensional EEG data matrix; Step 2, using the spatial filtering co-space mode to extract the variance features of two-dimensional motor imagery EEG from the high-dimensional EEG data matrix; Step 3, treating each feature corresponding to each experiment as A eigenvector, calculating the intra-class dispersion and inter-class dispersion of each eigenvector; step 4, according to the Fisher ratio principle, using the two eigenvectors corresponding to each experiment as the optimal EEG feature; step 5, Using semi-supervised interpretable clustering models, discriminant rectangular mixture models identify optimal EEG features across multiple subjects. This method is an effective EEG clustering method for single motor imagery. When a certain classification accuracy is guaranteed, no training set is needed and the training time is shortened.

Description

technical field [0001] The invention relates to the technical field of brain-computer interface, in particular to a binary classification motor imagery EEG signal recognition method based on an interpretable clustering model. Background technique [0002] In the field of brain-computer interface technology, one of the most widely used biological signals is the EEG signal generated by the neuron population in the human brain. In recent years, the brain-computer interface has become a popular research technology, which provides a bridge for people with severe motor dysfunction to communicate with external devices without relying on the neuromuscular communication system, such as developing brain function-controlled wheelchairs, arms and other devices. In addition, it can also help develop new smart home, smart entertainment and other industries. The neuropsychological features commonly used in the study of cerebral infarction mainly include motor imagery, steady-state visual ...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the problems of high training cost, long training time and low clustering accuracy in current EEG signal recognition

Images