The Sandwich enhanced Convolutional Block Attention Module (SCBAM) achieved an average accuracy of 78.63% in binary classification and 61.12% in five-class classification of individual finger movements using ultra-high-density EEG data.
The proposed SCBAM network significantly improves the accuracy of individual finger classification using ultra-high-density EEG data, offering potential for dexterous task decoding in Brain-Computer Interfaces.
p-value: p=0.0122
Introduction: Ultra-High-Density Electroencephalography (uHD EEG) has gained increasing attention for its potential in individual finger decoding. However, accurately classifying these movements remains challenging due to the subtle spatial overlaps in cortical activity, which standard architectures often fail to isolate. Methods: To address this, we propose the Sandwich enhanced Convolutional Block Attention Module (SCBAM). The unique sandwich structure integrates dual attention mechanisms between convolutional layers, enabling the network to more effectively refine high-dimensional spatial features. Results and discussion: The proposed network achieves an average accuracy of 78.63 (1.56)% in binary classification across ten finger pairs in five subjects, with the highest accuracy of 85% obtained at Thumb vs. Ring. The proposed network achieves an average accuracy of 61.12 (0.95)% in five-class classification across five subjects, with a highest accuracy of 62.36% on subject S2. The five-class classification is performed using 10 binary classifiers under a one-vs.-one strategy. Notably, five-class classification of individual fingers has not been extensively explored in the current literature, particularly with high-density EEG (HDEEG) data. This study addresses this gap, offering a valuable reference for future discussions. We conduct ablation studies to investigate the individual and synergistic effects of the modules in the proposed model. The results highlight the effects of two sequential attention mechanisms in this task. We conduct comparative experiments of our proposed model against six benchmark networks. The results from SCBAM significantly outperform these established models with FBCSP features. The proposed SCBAM significantly improves accuracy in binary finger classification compared to SVM and MLP using the same uHD EEG dataset. In summary, this study presents a high-performance hybrid network for individual finger classification and highlights the potential of uHD EEG for dexterous task decoding in Brain-Computer Interfaces (BCI).
Zhang et al. (Wed,) conducted a other in Healthy subjects (finger movement decoding) (n=5). Sandwich enhanced Convolutional Block Attention Module (SCBAM) vs. Benchmark networks (e.g., EEGNet, EEGLearn, Shallow ConvNets, Deep ConvNets) was evaluated on Average accuracy in binary classification across ten finger pairs (95% CI 77.26-80.00, p=0.0122). The Sandwich enhanced Convolutional Block Attention Module (SCBAM) achieved an average accuracy of 78.63% in binary classification and 61.12% in five-class classification of individual finger movements using ultra-high-density EEG data.