THRESHOLD ADAPTATION IN AUTOMATIC WAVELET-ICA FOR ELECTROENCEPHALOGRAPHIC ARTIFACT REMOVAL

Nadia Mammone, Aime’ Lay-Ekuakille, Patrizia Vergallo, Francesco C. Morabito

Abstract

Electroencephalography (EEG) is a well established methodology to record the electrical activity of the brain. We can be interested in monitoring the cerebral electrical activity for different purposes: studying the cognitive activity, interfacing the brain with the machine, extracting diagnostic information, etc. Artifacts are unwelcome signals, generated by electromagnetic sources not related to cerebral activity, that may overlap to the EEG signals and affect their processing. Whatever the goal of EEG processing, a preprocessing step consisting in artifact removal is normally required. Unfortunately, artifact removal is unavoidably a lossy procedure, therefore, the goal must be removing artifacts losing the minimum amount of useful information embedded in the EEG. To this purpose, Automatic Wavelet-ICA was recently proposed by the authors. The technique is multistep and parameter dependent, thus its performance may vary signi?cantly with the parameter setting. The present paper shows the results of the optimization with respect to the threshold used for artifact detection.

Keywords

automatic artifact rejection, electroencephalography, wavelet transform, independent component analysis, entropy

Download

IMEKO-TC13-2014-20.pdf

DOI

-

IMEKO TC

TC13 - Measurements in Biology and Medicine

Back to the proceedings