Semi-supervised EEG emotion recognition model based on enhanced graph fusion and GCN

Abstract

Objective. To take full advantage of both labeled data and unlabeled ones, the Graph Convolutional Network (GCN) was introduced in electroencephalography (EEG) based emotion recognition to achieve feature propagation. However, a single feature cannot represent the emotional state entirely and precisely due to the instability of the EEG signal and the complexity of the emotional state. In addition, the noise existing in the graph may affect the performance greatly. To solve these problems, it was necessary to introduce feature/similarity fusion and noise reduction strategies. Approach. A semi-supervised EEG emotion recognition model combining graph fusion, network enhancement, and feature fusion was proposed. Firstly, different features were extracted from EEG and then compacted by Principal Component Analysis (PCA), respectively. Secondly, a Sample-by-sample Similarity Matrix (SSM) was constructed based on each feature, and Similarity Network Fusion (SNF) was adopted to fuse the graphs corresponding to different SSMs to take advantage of their complementarity. Then, Network Enhancement (NE) was performed on the fused graph to reduce the noise in it. Finally, GCN was performed on the concatenated features and the enhanced fused graph to achieve feature propagation. Main results. Experimental results demonstrated that: i) When 5.30% of SEED and 7.20% of SEED-IV samples were chosen as the labeled samples, respectively, the minimum classification accuracy improvement achieved by the proposed scheme over state-of-the-art schemes were 1.52% on SEED and 13.14% on SEED-IV, respectively. ii) When 8.00% of SEED and 9.60% of SEED-IV samples were chosen as the labeled samples, respectively, the minimum training time reduction achieved by the proposed scheme over state-of-the-art schemes were 46.75s and 22.55s, respectively. iii) Graph fusion, network enhancement, and feature fusion all contributed to the performance enhancement. iv) The key hyperparameters that affect the performance were relatively few and easy to set to obtain outstanding performance. Significance. This paper demonstrated that the combination of graph fusion, network enhancement, and feature fusion help to enhance GCN-based EEG emotion recognition.

MSLTE: multiple self-supervised learning tasks for enhancing EEG emotion recognition

“…In ECLGCNN [18], GCN is combined with LSTM to extract spatial-temporal features from EEG for emotion recognition. In [19], the graph fusion and graph enhancement are integrated into GCN to construct a semi-supervised EEG emotion recognition model called EGFG. In [20], a spatial-temporal attention mechanism and a self-adaptive brain network adjacency matrix are designed to capture the significant sequential segments and spatial location information in EEG signals, and aim to represent the diverse activation patterns under different emotion categories.…”

Decoding auditory-evoked response in affective states using wearable around-ear EEG system

“…Most commonly used systems involve electrodes positioned over the scalp based on the international 10-20 systems. Several works demonstrated that high-quality EEG signals measured from the scalp can effectively be used for emotion recognition [18,[22][23][24][25]. However, conventional scalp-based EEG acquisition methods are uncomfortable to use over a long period of time, which may lead to frustration and evoke negative emotions over the session.…”

Adaptive Gated Graph Convolutional Network for Explainable Diagnosis of Alzheimer’s Disease Using EEG Data

“…The design of GNN encoders in step (2) for EEG applications has been mainly limited to simple architectures, such as the Chebyshev graph convolution (ChebConv) [30], [31], [33], [35]- [37], [40], and simple graph convolution (GCN) [8], [29], [34], [41]- [43]. However, we hypothesise that such node embedding updating mechanisms are not optimal for EEG tasks.…”

“…Node features are commonly defined as EEG time-series signal [11], [29]- [31], or a statistical summary of the signal in the time domain [32], [33], the frequency domain [8], [34], or the differential entropy [29], [34]- [38]. Based on network neuroscience literature, many approaches define the brain graph using FC measures [8], [11], [29], [31]- [33], [39], [40]. The graph structure can also be based on the distance between EEG electrodes [33], [35], [36].…”