Efficient Multi-View Graph Convolutional Network with Self-Attention for Multi-Class Motor Imagery Decoding

• Published in: Bioengineering (Basel) Vol. 11; no. 9; p. 926
• Main Authors: Tan, Xiyue, Wang, Dan, Xu, Meng, Chen, Jiaming, Wu, Shuhan
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 01.09.2024; MDPI
• Subjects: Accuracy; Analysis; Artificial neural networks; Attention; Biochips; Brain; Brain research; brain–computer interface; Classification; Computer applications; Datasets; Deep learning; EEG; Electrodes; Electroencephalography; Euclidean space; Feature extraction; Feature selection; graph convolutional networks; Graph representations; Human-computer interface; Imagery; Implants; Machine learning; Medical research; Mental task performance; motor imagery; Neural networks; Receptive field; self-attention; Signal classification; Topology; deep learning; brain–computer interface; motor imagery; self-attention; graph convolutional networks
• ISSN: 2306-5354; 2306-5354
• DOI: 10.3390/bioengineering11090926

Research on electroencephalogram-based motor imagery (MI-EEG) can identify the limbs of subjects that generate motor imagination by decoding EEG signals, which is an important issue in the field of brain–computer interface (BCI). Existing deep-learning-based classification methods have not been able to entirely employ the topological information among brain regions, and thus, the classification performance needs further improving. In this paper, we propose a multi-view graph convolutional attention network (MGCANet) with residual learning structure for multi-class MI decoding. Specifically, we design a multi-view graph convolution spatial feature extraction method based on the topological relationship of brain regions to achieve more comprehensive information aggregation. During the modeling, we build an adaptive weight fusion (Awf) module to adaptively merge feature from different brain views to improve classification accuracy. In addition, the self-attention mechanism is introduced for feature selection to expand the receptive field of EEG signals to global dependence and enhance the expression of important features. The proposed model is experimentally evaluated on two public MI datasets and achieved a mean accuracy of 78.26% (BCIC IV 2a dataset) and 73.68% (OpenBMI dataset), which significantly outperforms representative comparative methods in classification accuracy. Comprehensive experiment results verify the effectiveness of our proposed method, which can provide novel perspectives for MI decoding.