Using SSVEP based brain-computer interface to control functional electrical stimulation training system

In this work, a functional electrical stimulation (FES) training system using steady-state visual evoked potential (SSVEP) based brain-computer interface (BCI) was designed to realize the control of upper limb movements. Subjects were required to initiatively focus on one of five flickering lights with different frequencies on the computer screen, while the electroencephalogram (EEG) signal was acquired from the channels at the visual cortex region. The five primary flickering frequencies and their harmonic components were extracted as classification features from the EEG channels at the visual cortex region, and then linear discriminant analysis (LDA) classifier in pairwise strategy was used to decode the subject's intention corresponding to the flickering light that the subject was focusing on. Thereafter the user's intention was transformed into a command to trigger the FES system to generate the desired stimulation pattern. The experimental results showed that the feature extraction and classification methods were efficient in on-line classification. Moreover an energy bar was applied to the human-machine interaction interface to enhance the performance of the system as a dynamic feedback to the user. The results indicated that the subjects could control the FES training system to realize the predefined action sequences with their own intention.