Multi-task EEG Signal Classification using Correlation-based IMF Selection and Multi-class CSP

In the context of motor imagery (MI)-based brain-computer interface (BCI) systems, a great amount research has been studied for attaining higher classification performance by extracting discriminative features from MI-based electroencephalogram (EEG) signals. this study, we propose an innovative approach classifying multi-class MI-EEG signals, which consists signal processing technique based on empirical mode decomposition (EMD) and common spatial patterns (MCCSP). Specifically, after applying EMD, selecting best intrinsic functions (IMF) as substitution to original EEG next stage processing. The metric used selection is cross-correlation each decomposed IMF with signal. Next, extend CSP algorithm MCCSP be utilized feature extractor. We applied our BCI competition IV (2a). Results revealed that proposed improved accuracy significantly compared case when directly channel data. Moreover, K-nearest neighbor (KNN) achieved highest mean rate 91.28%. Our findings suggest promising elevated 96.71% can raising dimension through MCCSP. Compared state-of-the-art algorithms, method highly convincing motivating future studies.