A method for automatic removal of eye blink artifacts from EEG based on EMD-ICA

Abstract

The electroencephalography (EEG) recordings are mostly contaminated by eye blink artifacts. It is very difficult to analyze and interpret the EEG signal due to frequent occurrence of the eye blink artifact. In this paper, a new hybrid algorithm that automatically removes the eye blink artifact from the EEG, based on Empirical Mode Decomposition (EMD) and Independent Component Analysis (ICA) is proposed. The proposed algorithm is evaluated on simulated EEG to calculate correlation coefficient and signal-to-artifact ratio (SAR). A non-corrected EEG was simulated to have a SAR of -19.1673 dB. From the simulation results, the highest average correlation coefficient and SAR of corrected EEG from non-corrected EEG are obtained as 0.871094 and 2.71645 dB respectively by applying proposed algorithm. The results demonstrate that proposed method recovers the EEG data by removing the eye blink artifacts reliably. In addition, the proposed method is applied on real spontaneous EEG data with eye blink artifact.