Multi-Direction Decoding of Both-Hand Movement Using EEG Signals

In this paper, we propose a method to decode the both-hand movement multi-direction based on electroencephalogram (EEG) signals. We use two kinds of decoding features, which are the potential amplitudes and power sums of EEG signals. One-versus-rest and decision tree are adopted as classification strategies, and linear discriminant analysis (LDA) classifier is used for classification. We apply an experimental paradigm to demonstrate the proposed method. The best four-class classification performance using the power sums of EEG signals with the one-versus-rest classification strategy is close to 70%. The experimental results show the feasibility of decoding both-hand movement multi-directions based on EEG signals. This work can promote the development of brain-computer interfaces for the assistance of hand-impaired patients.