A multi-head residual connection GCN for EEG emotion recognition

• Published in: Computers in biology and medicine Vol. 163; p. 107126
• Main Authors: Qiu, Xiangkai, Wang, Shenglin, Wang, Ruqing, Zhang, Yiling, Huang, Liya
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.09.2023; Elsevier Limited
• Subjects: Accuracy; Artificial neural networks; Brain; Classification; Complex brain network; EEG; Electrodes; Electroencephalography; Emotion recognition; Emotions; Entropy; Graph convolutional neural network; Graph neural networks; Internal Medicine; Neural networks; Other; Stability augmentation; Emotion recognition; Complex brain network; Graph convolutional neural network; EEG
• ISSN: 0010-4825; 1879-0534; 1879-0534
• DOI: 10.1016/j.compbiomed.2023.107126

Electroencephalography (EEG) emotion recognition is a crucial aspect of human-computer interaction. However, conventional neural networks have limitations in extracting profound EEG emotional features. This paper introduces a novel multi-head residual graph convolutional neural network (MRGCN) model that incorporates complex brain networks and graph convolution networks. The decomposition of multi-band differential entropy (DE) features exposes the temporal intricacy of emotion-linked brain activity, and the combination of short and long-distance brain networks can explore complex topological characteristics. Moreover, the residual-based architecture not only enhances performance but also augments classification stability across subjects. The visualization of brain network connectivity offers a practical technique for investigating emotional regulation mechanisms. The MRGCN model exhibits average classification accuracies of 95.8% and 98.9% for the DEAP and SEED datasets, respectively, highlighting its excellent performance and robustness. •MRGCN model incorporates complex brain networks and graph neural networks (GNN) for profound EEG emotion recognition.•Use differential entropy to extract the complexity of EEG signals and to analyze the inner workings of emotion generation.•We designed a long-distance and short-distance brain network to explore complex topological characteristics.•Add residual-based architecture enhances performance and classification stability.•Visual optimal solution model was used to study connection mechanism among brain regions during production of emotions.