Ocular Artifacts Removal from EEG Using EMD

Electroencephalogram (EEG) provides a non-invasive way to analyze brain activity. Blinking and movement of the eyes causes a strong electrical activity that can contaminate EEG recordings, particularly around the forehead but also as far as in occipital areas. Removal of such ocular artifacts is a considerable signal processing problem, since those artifacts overlap in frequency domain with EEG. In this paper we propose a signal reconstruction method based on a time frequency analysis tool known as the Hilbert-Huang spectrum. We demonstrate how our reconstruction scheme can be successfully applied to contaminated EEG data for the purposes of removing unwanted ocular artifacts.