Graph convolutional neural network (GCNN) based methods have been widely used in electroencephalogram(EEG) related works due to their advantages of considering the symmetrical connections of brain regions. However, the current GCNN based methods do not fully explore other correlations between EEG channels. Many studies have proved that definite causal connections exist between brain regions. Therefore, this paper proposes a Causal Graph Convolutional Neural Network (CGCNN) using Granger causality test to calculate inter-channel interactions. First, we consider causal relations between EEG channels and construct an asymmetric causal graph with direction. Then, we adopt depthwise separable convolution to extract emotional features from multi-channel EEG signals. Experiments carried out on SEED and SEED-IV show that CGCNN has the ability to represent the causal information flow in different emotional states, and improve the classification accuracy to 93.36% on SEED and 75.48% on SEED-IV respectively. The results outperform other existing methods, indicating that Granger causality is more effective in revealing the correlations between EEG channels in emotion recognition.

Various relations existing in Electroencephalogram (EEG) data are significant for EEG feature representation. Thus, studies on the graph-based method focus on extracting relevancy between EEG channels. The shortcoming of existing graph studies is that they only consider a single relationship of EEG electrodes, which results an incomprehensive representation of EEG data and relatively low accuracy of emotion recognition. In this paper, we propose a fusion graph convolutional network (FGCN) to extract various relations existing in EEG data and fuse these extracted relations to represent EEG data more comprehensively for emotion recognition. First, the FGCN mines brain connection features on topology, causality, and function. Then, we propose a local fusion strategy to fuse these three graphs to fully utilize the valuable channels with strong topological, causal, and functional relations. Finally, the graph convolutional neural network is adopted to represent EEG data for emotion recognition better. Experiments on SEED and SEED-IV demonstrate that fusing different relation graphs are effective for improving the ability in emotion recognition. Furthermore, the emotion recognition accuracy of 3-class and 4-class is higher than that of other state-of-the-art methods.

Graph convolutional network (GCN) based on the brain network has been widely used for EEG emotion recognition. However, most studies train their models directly without considering network dimensionality reduction beforehand. In fact, some nodes and edges are invalid information or even interference information for the current task. It is necessary to reduce the network dimension and extract the core network. To address the problem of extracting and utilizing the core network, a core network extraction model (CWGCN) based on channel weighting and graph convolutional network and a graph convolutional network model (CCSR-GCN) based on channel convolution and style-based recalibration for emotion recognition have been proposed. The CWGCN model automatically extracts the core network and the channel importance parameter in a data-driven manner. The CCSR-GCN model innovatively uses the output information of the CWGCN model to identify the emotion state. The experimental results on SEED show that: (1) the core network extraction can help improve the performance of the GCN model; (2) the models of CWGCN and CCSR-GCN achieve better results than the currently popular methods. The idea and its implementation in this paper provide a novel and successful perspective for the application of GCN in brain network analysis of other specific tasks. The code is available at https://github.com/ykhdu/CWGCN-CCSR-GCN .

Electroencephalography (EEG) signals may provide abundant information reflecting the developmental changes in brain status. It usually takes a long time to finally judge whether a brain is dead, so an effective pre-test of brain states method is needed. In this paper, we present a hybrid processing pipeline to differentiate brain death and coma patients based on canonical correlation analysis (CCA) of power spectral density, complexity features, and feature fusion for group analysis. In addition, time-varying power spectrum and complexity were observed based on the analysis of individual patients, which can be used to monitor the change of brain status over time. Results showed three major differences between brain death and coma groups of EEG signal: slowing, increased complexity, and the improvement on classification accuracy with feature fusion. To the best of our knowledge, this is the first scheme for joint general analysis and time-varying state monitoring. Delta-band relative power spectrum density and permutation entropy could effectively be regarded as potential features of discrimination analysis on brain death and coma patients.

Electroencephalography (EEG)-based motor imagery (MI) is one of brain computer interface (BCI) paradigms, which aims to build a direct communication pathway between human brain and external devices by decoding the brain activities. In a traditional way, MI BCI replies on a single brain, which suffers from the limitations, such as low accuracy and weak stability. To alleviate these limitations, multi-brain BCI has emerged based on the integration of multiple individuals’ intelligence. Nevertheless, the existing decoding methods mainly use linear averaging or feature integration learning from multi-brain EEG data, and do not effectively utilize coupling relationship features, resulting in undesired decoding accuracy. To overcome these challenges, we proposed an EEG-based multi-brain MI decoding method, which utilizes coupling feature extraction and few-shot learning to capture coupling relationship features among multi-brains with only limited EEG data. We performed an experiment to collect EEG data from multiple persons who engaged in the same task simultaneously and compared the methods on the collected data. The comparison results showed that our proposed method improved the performance by 14.23% compared to the single-brain mode in the 10-shot three-class decoding task. It demonstrated the effectiveness of the proposed method and usability of the method in the context of only small amount of EEG data available.