Automatic removal of ocular artifacts from EEG data using adaptive filtering and independent component analysis

Abstract

A method to eliminate eye movement artifacts based on Independent Component Analysis (ICA) and Recursive Least Squares (RLS) is presented. The proposed algorithm combines the effective ICA capacity of separating artifacts from brain waves, together with the online interference cancellation achieved by adaptive filtering. The method uses separate electrodes localized close to the eyes (Fp1, Fp2, F7 and F8), that register vertical and horizontal eye movements, to extract a reference signal. Each reference input is first projected into ICA domain and then the interference is estimated using the RLS algorithm. This interference estimation is subtracted from the EEG components in the ICA domain. Results from experimental data demonstrate that this approach is suitable for eliminating artifacts caused by eye movements, and the principles of this method can be extended to certain other sources of artifacts as well. The method is easy to implement, stable, and presents a low computational cost.