MS-HyFS: A novel multiscale hybrid framework with Scalable electrodes for motor imagery classification

Ziheng Guo, Yuan Feng, Ming Ma, Yudi Sun, Likun Xia

Abstract: Highlights•A feature extraction method that encapsulate temporal, spatial, local, and global features within EEG signals.•An improved feature extraction method that captures multiscale and multi-domain features with overlapping time windows.•An optimal electrode placement to reduce the computational cost based on asymmetric properties and neural activity areas of the brain.•Significantly enhance the classification performance with higher average kappa and lower standard deviation of different subjects.