Removing Electroencephalographic Artifacts by Independent Components Analysis

Abstract

Pervasive electroencephalographic (EEG) artifacts associated with blinks, eye-movements, muscle noise, cardiac signals, and line noise poses a major challenge for EEG interpretation and analysis. Many methods have been proposed to remove artifacts from EEG recordings, especially those arising from eye movements and blinks. In Vietnam, epilepsy analysis methods, based on the EEG, is still the visual inspection of the EEG by a highly skilled electroencephalographer or neurophysiologist. But the doctors meet with the difficulties when the obtained scalp EEG recording have not only brain activities sginals but also the artifacts. Here, we use the independent components analysis (ICA) method to separate the clean data from the rest of the sources, artifacts within the brain. The separated independent components are good, we will clear the indpendent components is not brain activity sources and get EEG signals without the artifacts.