A VR Combined with MI-BCI Application for Upper Limb Rehabilitation of Stroke

Abstract

Virtual Reality (VR) is expected to significantly improve the motor function for stroke as a rehabilitation intervention technique. VR can provide a simulated world comparable to the real world that allows clients to interact with the virtual environment (VE), and provides real-time feedback. The principle of rehabilitation method based on motor imagery (MI) brain-computer interface (BCI) is to train the motor nerves directly through MI to promote the plasticity of the central nervous system. To rehabilitate the upper limb function of stroke patients, a rehabilitation system based on VR and MI-BCI is developed in this paper, which integrates the real-life scenario into the monotonous rehabilitation training. The rehabilitation system provides 6 training scenarios and 9 training movements. The system will provide real-time visual, auditory and haptic multisensory-feedback, which makes training more interesting, promotes motor learning, and enhances participation in a rehabilitation process. An experiment was conducted with healthy people by selecting 4 groups of training scenes with different characteristics from the rehabilitation system, and preliminarily verified the system can give positive guidance to people as to motion imagination. Therefore, the system combining MI-BCI and VR is expected to help stroke patients produce more recognizable EEG signals, and achieved self-initiative rehabilitation.