SSVEP based brain-computer interface controlled functional electrical stimulation system for upper extremity rehabilitation

Brain-computer interface (BCI) is currently developed as an alternative technology with a potential to restore lost motor function in patients with neurological injuries. In this paper, we describe an integrated system of a non-invasive electroencephalogram (EEG)-based BCI with a non-invasive functional electrical stimulation (FES). This system enables the direct brain control of upper limbs to achieve motor rehabilitation. The EEG signals based on steady-state visual evoked potential (SSVEP) were used in the BCI. The classifier of linear discriminant analysis was applied to deal with the frequency domain characteristics and recognize intentions. The identified intentions were transformed into instructions to trigger FES which was controlled with iterative learning control method to stimulate the relevant muscles of upper limbs for motor recovery. Results show that the integration of BCI with an upper-extremity FES is feasible with an average accuracy of about 73.9% over five able-bodied subjects.