Comparison Between Independent Component Analysis and Wiener-Hopf Filter Techniques for Eye Blink Removal

Electroencephalogram (EEG) is useful in the clinical diagnosis of various brain diseases. However, the existence of artifacts corrupting the EEG reduces the ability to do such a proper diagnosis. One of these unwanted signals is the eye blink. Therefore, pre-processing of the EEG is of great importance for proper diagnosis. This work compares between using the Independent component Analysis (ICA) and the Wiener filter techniques for removing eye blink (EB) artifacts from EEG. EEG datasets contaminated with EBs are used for quantifying the performance of the EB rejection methods. Mean squared error (MSE) is calculated and compared between the two methods. The Results show that the Wiener-Hopf filter gives better results in rejecting EBs and producing artifact-free EEG signal.