Graph Neural Network-based EEG Classification: A Survey

TL;DR

This survey reviews the use of graph neural networks for EEG classification, categorizing methods, identifying trends like spectral convolution, and discussing future research directions such as transfer learning.

Contribution

It provides a systematic review and categorization of GNN-based EEG classification methods, highlighting prevalent techniques and future research opportunities.

Findings

Spectral graph convolutional layers are more common than spatial.

Raw EEG signals and differential entropy are popular node features.

Emerging trends include transfer learning and cross-frequency interaction modeling.

Abstract

Graph neural networks (GNN) are increasingly used to classify EEG for tasks such as emotion recognition, motor imagery and neurological diseases and disorders. A wide range of methods have been proposed to design GNN-based classifiers. Therefore, there is a need for a systematic review and categorisation of these approaches. We exhaustively search the published literature on this topic and derive several categories for comparison. These categories highlight the similarities and differences among the methods. The results suggest a prevalence of spectral graph convolutional layers over spatial. Additionally, we identify standard forms of node features, with the most popular being the raw EEG signal and differential entropy. Our results summarise the emerging trends in GNN-based approaches for EEG classification. Finally, we discuss several promising research directions, such as exploring the potential of transfer learning methods and appropriate modelling of cross-frequency interactions.