Early diagnosis of Alzheimer disease using EEG signals: the role of pre-processing

Abstract

Electroencephalograms (EEGs) have significant ability to measure the brain activity and have huge potential for the analysis of the brain diseases like Alzheimer disease (AD). EEG is a measurement of electrical signal generated from the neurons presents in the brain. These non-stationary EEGs signals show the sign of many current diseases or even give the warning about impending diseases. Three main effects of Alzheimer disease on EEG signal have been identified like signal slowing, reduction in EEG complexity and a change in the normal state of EEG synchrony. Brain computer interface (BCI) system gives a way for the detection of the preliminary stage of the Alzheimer disease based on nonlinear EEG signals. Pre-processing of the EEG decides the efficiency of this methodology. Artefacts must be removed before analysing the EEG signals. Henceforth in recent year, pre-processing of EEG signals has got a great deal of enthusiasm for researchers. In this paper, state of art EEG pre-processing techniques is explored. This paper indicates clear and simple understanding of selected pre-processing techniques with respect to Alzheimer disease diagnosis.