Electroencephalogram signal classification method based on visual Transform

Abstract

The invention discloses an electroencephalogram signal classification method based on vision Transform, which comprises the following steps: firstly, carrying out data preprocessing to obtain processed EEG (electroencephalogram) data with labels, and then constructing an electroencephalogram signal classification model based on the vision Transform; and finally, training an electroencephalogram signal classification model through the preprocessed EEG data. According to the method, the EEG samples are subjected to feature embedding through a proper EEG feature embedding method, then the long-time dependency relationship between the local features of the EEG samples and the continuous electroencephalogram signals is learned, and better performance is obtained in an electroencephalogram signal classification task.

Description

technical field [0001] The invention relates to the field of electroencephalogram signal recognition in the field of biological feature recognition, in particular to a method for classifying electroencephalogram signals based on a visual transformer (Vision Transformer, ViT). Background technique [0002] Electroencephalogram (Electroencephalogram, EEG) is a method that uses electrophysiological indicators to record brain activity. When the brain is active, the synchronous post-synaptic potentials of a large number of neurons are summed and formed. It records the electric wave changes during brain activity, which is the overall reflection of the electrophysiological activities of brain nerve cells on the surface of the cerebral cortex or scalp. EEG signals contain rich, diverse and objective physiological information, and their research and analysis are often used in Brain Computer Interface (BCI) to realize the direct connection between the human or animal brain and externa...

AI Technical Summary

Problems solved by technology

[0005] In order to overcome the above-mentioned poor performance of applying the transformer in EEG signal classification and the inability to learn local features of features, the present invention proposes a visual Transformer-based EEG signal classification method EEGVision Transformer (EEGViT). The EEG Transformer Encoder module learns local features, and the Sequence In Time Transformer Encoder learns the timing dependencies between continuous samples to improve classification performance

Images