ESSDM: An Enhanced Sparse Swarm Decomposition Method and Its Application in Multi-Class Motor Imagery-Based EEG-BCI System

Electroencephalogram (EEG)-based motor imagery (MI) (MI-EEG) decoding has established a novel experimental paradigm in brain-computer interface (BCI) applications that offer effective treatment for stroke paralyzed patients. However, existing MI-EEG-based BCI systems introduce deployment issues because of nonstationary EEG signals, suboptimal features, and limited multiclass scalability. To tackle these issues, we propose an enhanced sparse swarm decomposition method (ESSDM) based on selfish-herd optimization and sparse Fourier transform (FT) spectrum to solve the issue of choice of uniform decomposition and hyperparameters in swarm decomposition, and further applied to enhance MI-EEG classification. ESSDM adopts improved swarm filtering to automatically deliver optimal frequency bands in the sparse FT spectrum with optimized hyperparameters to extract dominant oscillatory components (OCs) that significantly enhance MI activation-related sub-bands. In addition, new fitness criteria has been designed based on the Kullback-Leibler divergence distance from the spectral kurtosis of the obtained modes to select hyperparameters that optimize decomposition effect, avoid excessive iterations, and provide fast convergence with optimal modes. Further, fused time-frequency graph (FTFG) features have been derived from computed time-frequency representation to find cross-channel mutual spectral information. The experimental results on the 2-class BCI III-4a, 2-class OpenBMI, and 4-class BCI IV-2a datasets reveal that the proposed framework based on FTFG features with capsule neural network (CapsNet), ESSDM-FTFG-CapsNet outperformed other existing methods in specific-subject, cross-subject, and cross-session scenarios. Note to Practitioners—This paper addresses the challenges in developing practical BCI systems using MI-EEG signals, specifically focusing on the accurate extraction of OCs corresponding to MI activation patterns from any multichannel MI-EEG signals and the subsequent design of optimal features for automated MI-EEG classification. This paper suggests a novel decomposition technique to extract optimal OCs for efficient feature extraction, and an automated BCI framework is further developed for multiclass MI tasks classification by considering practical testing scenarios. The proposed BCI frameworks enhance MI-EEG task detection accuracy and demonstrates promising improvement in practical BCI system for clinical practices. However, the presented BCI system approach involve limited number of MI-EEG task classes in the cross-subject, cross-dataset, and cross-session validation. In future studies, the presented approach can be studied on more number of MI-EEG task problems in diverse cross-day datasets for achieving generalized MI-EEG BCI model.