Electroencephalogram signal analysis method based on graph convolutional network

Abstract

The present invention relates to an electroencephalogram signal analysis method. The method includes the following steps: S1, establishing a brain functional network based on electroencephalogram signals; S2, labelling different brain functional networks according to the brain functional network to complete graph convolution network model training; and S3, performing type recognition on the novelbrain functional network according to the trained graph convolutional network model to complete electroencephalogram classification. Compared with the prior art, the electroencephalogram signal analysis method provided by the invention can accurately describe the detailed characteristics and state of a brain, the method can provide a novel method for intelligent judgment, early warning and treatment of the diseases when applied to the brain functional network of a patient with cerebral nervous system related diseases.

Description

technical field [0001] The invention belongs to the field of electroencephalogram signal analysis, and in particular relates to an electroencephalogram signal analysis method based on a graph convolutional network. Background technique [0002] Electroencephalogram (EEG) is a graph obtained by amplifying and recording the spontaneous biopotential of the brain from the scalp through sophisticated electronic instruments. It is the spontaneous and rhythmic electrical activity of brain cell groups recorded by electrodes. EEG signals have high time precision and can dynamically observe changes in the state of the brain, so EEG signals are an important tool for diagnosing different neurological disorders and diseases. The interaction and interconnection between neurons in the brain make the brain a complex dynamical system. Viewing the brain as a functional network and studying the temporal correlation between neural clusters in different areas of the brain to explore the connect...

AI Technical Summary

Problems solved by technology

At present, the establishment method of brain functional network is generally to establish a functional network based on the Pearson correlation coefficient of EEG signals collected between different electrodes, or to establish a functional network based on the phase-locked value of EEG signals between different electrodes, which ignores the The discharge characteristics of EEG signals at different moments, the established brain function network cannot better reflect the real state of the brain

In addition, traditional EEG signal analysis methods generally analyze the differences in EEG signals between patients with diseases related to the brain nervous system and normal people from the characteristics of the signal itself (amplitude, frequency spectrum, phase), and cannot analyze the overall connectivity of the brain. signal, thus ignoring some high-order hidden valuable information of the brain network

Images