Classification of Artifacts in EEG Signal Recordings

Abstract

The EEG (electroencephalographic) recordings determine the electric driving forces produced in the brain, in reaction to the given stimulus. The unconstrained EEG information is utilized for conclusion and treatment of some brain ailments/ diseases. For the information to be utilized for clinical applications, it should be liberated from the different artifacts like the eye blinks, movements, head movements as well as muscle activity. These artifacts should be rectified or the influenced parts should be evacuated in the pre-processing of the EEG dataset. With enormous number of datasets to be investigated, it is important to have consistency in the examination. Uniformity, reproducibility and reliability in the pre-processed information can be acquired if a statistical approach is taken while pre-processing the datasets. In a perfect world, this can be semi-or completely automated. This loom therefore, should be taken while eradicating the less frequently occurring artifacts and correcting the more frequently occurring artifacts, in order to retain more complete datasets for additional research or clinical purposes. In this paper broad classification of EEG artifacts and methods to detect and its removal is discussed.