Single trial method for brain-computer interface

Electroencephalogram (EEG) related to fast eye movement (saccade), has been the subject of application oriented research by our group toward developing a brain-computer interface (BCI). Our goal is to develop novel BCI based on eye movements system employing EEG signals online. Most of the analysis of the saccade-related EEG data has been performed using ensemble averaging approaches. However, ensemble averaging is not suitable for BCI. In order to process raw EEG data in real time, we performed saccade-related EEG experiments and processed data by using the non-conventional fast ICA with reference signal (FICAR). The FICAR algorithm can extract desired independent components (IC) which have strong correlation against a reference signal. Visually guided saccade tasks and auditory guided saccade tasks were performed and the EEG signal generated in the saccade was recorded. The EEG processing was performed in three stages: PCA preprocessing and noise reduction, extraction of the desired IC using Wiener filter with reference signal, and post-processing using higher order statistics fast ICA based on maximization of kurtosis. Form the experimental results and analysis we found that using FICAR it is possible to extract form raw EEG data the saccade-related ICs and to predict saccade in advance by about 10 [ms] before real movements of eyes occurs. For single trail EEG data we have successfully extracted the desire ICs with recognition rate about 70%. In next steps, saccade-related EEG signals and saccade-related ICs in visually and auditory guided saccade task are compared in the point of the latency between starting time of a saccade and time when a saccade-related EEG signal or an IC has maximum value and in the point of the peak scale where a saccade-related EEG signal or an IC has maximum value. As results, peak time when saccade-related ICs have maximum amplitude is earlier than peak time when saccade-related EEG signals have maximum amplitude. This is very important advantage for developing our BCI. However, S/N ratio in being processed by FICAR is not improved comparing S/N ratio in being processed by ensemble averaging.