A Hybrid Brain Computer Interface based on SSVEP And Eye Tracking

Abstract

Brain-computer interface (BCI) had extensive application prospect in medical rehabilitation and other fields. Steady-state visual evoked potential (SSVEP) was one of the most widely used BCIs. But SSVEP had disadvantages such as small instruction set and low information transfer rates. In order to solve the problems, this paper proposed a high- performance hybrid brain-computer interface that combined steady-state visual evoked potentials and eye tracking, which used only six frequencies to increase the number of instruction sets to 48. The task related component analysis (TRCA) algorithm was used for SSVEP classification. The 48-targets classification was achieved by combining the eye tracking classification results. Nine subjects participated in the offline experiment. The results showed that the average classification accuracy and information transfer rates of the hybrid BCI system combined with SSVEP and eye tracking were 89.00% and 136.96bit/min, respectively under the data length of 2S. The proposed system performed better than single modal BCI.