SSVEP based brain-computer interface controlled functional electrical stimulation system for knee joint movement

Abstract

Knee joint movement control plays an important part in the rehabilitation of lower limb locomotion. In this work, a functional electrical stimulation (FES) training system utilizing steady-state visual evoked potential (SSVEP) based brain-computer interface (BCI) was designed to realize the functional control of knee movement, including selection of multiple patterns: left or right knee joint for stimulation, start or stop of knee movement, acceleration or deceleration of knee movement. In order to investigate the problem of FES artifact on EEG signals, two classifiers were trained under normal condition (electrical artifact free) and artifact condition (initiatively adding FES artifact) respectively. The classifiers were used in pre-stimulation and on-stimulation separately after training. During the experiment, the subject was required to realize the predefined experiment paradigm, during which both the knee joint angle and real-time SSVEP selection were recorded to evaluate the coordination of BCI and FES integration. The experiment result showed the necessity and effectiveness of the proposed real-time classifier strategy, and knee joint movement control was successfully realized by subjects intention. © Springer-Verlag Berlin Heidelberg 2012.