Bimanual Arm Movements Decoding using Hybrid Method

Abstract

In arm movement BCI (brain-computer interface), the unimanual research has been well. However, the bimanual brain state is known to be different from the unimanual one, so the conventional arm movement decoding method seems to be insufficient to decode bimanual movement. In this research, we suggested the hybrid method to improve the decoding accuracy for bimanual movement estimation. The method consists of two step; 1st step: the movement conditions classification, and 2nd step: the hand trajectory prediction algorithm. As a result, the hybrid method showed improved arm movement decoding performance and significant and stable decoding rate over several months for bimanual tasks. This technique could be applied to arm movement BCI in real world and the various neuro-prosthetics fields.