Three-dimensional upper limb movement decoding from EEG signals

Abstract

A brain-computer interface (BCI) can be used to control a limb neuroprosthesis in patients. In particular, decoding trajectory of upper limb with motor imagery (MI) can support motor rehabilitation using a wearable robotic arm. Recent research shows the possibility of decoding hand movement trajectory from electroencephalography (EEG) signals. However, such studies are insufficient to apply motor rehabilitation, which are only considered hand movement trajectory. Although disabilities patients take correct hand movement, sometimes wrong elbow movement can be taken in motor rehabilitation. In this study, we explore to decode velocity of both hand and elbow at the same time from EEG signals when subjects move upper limb. The result shows feasibility toward controlling robotic arm.