Decoding of Individual Finger Movement on One Hand Using Ultra high-density EEG

Abstract

Brain-computer interfaces (BCIs) can enable people with severe motor disorders to perform daily activities by controlling prosthetic devices. Notably, many activities involve complex movements based on the control of individual fingers. However, it is still a great challenge for traditional scalp EEG to decode individual finger movements intention on the one hand due to its low spatial resolution and poor signal-to-noise ratio. In this paper, we designed an ultra-high-density (UHD) scalp EEG system to acquire high-spatial-resolution information to decode individual finger movement on one hand. We further compared the classification performance of the UHD EEG and lower-density EEG settings. The results indicated that UHD EEG achieved an average classification accuracy of 80.86%, 66.19%, 50.18%, and 41.57% for the two-class, three-class, four-class, and five-class finger movement tasks, respectively. The results obtained by UHD EEG were significantly better than those obtained by the lower-density EEG. This study offers a promising approach for detecting individual finger movement, and it is meaningful for the development of high-density EEG-based BCIs.