Virtual hand rehabilitation with force guidance adaptable to mental states using brain-computer interface

Abstract

This paper presents a virtual hand dexterity training system where force guidance is activated autonomously in real time depending on the mental states of user, which are determined by electroencephalogram (EEG) signals acquired using brain-computer interface (BCI). In the context of handwriting, when EEG signals suggests a certain level of frustration, assistive forces are produced as an aid to completing the desired strokes. The study provides insights toward the development of intelligent rehabilitation system with automated guidance and natural user interface for self-paced practice.