Mutual information analysis on MCI-AD EEG signal during resting and task conditions

Abstract

Alzheimer's is a rapidly progressing disease that destroys brain nerve cells. Hence, there is a need for an early diagnosis at its onset. Electroencephalography (EEG) is an inexpensive biomarker for screening AD at its prodromal stages. The aim of this paper is to analyze the information transmission between distinct brain regions in mild cognitive impairment-Alzheimer's disease (MCI-AD) patients and healthy controls using Mutual Information (MI) analysis employing EEG signals. MI is subdivided into Auto Mutual Information (AMI) and Cross Mutual Information (CMI). EEG signal analysis of the healthy controls and MCI-AD patients is carried out under resting and mental arithmetic task conditions. Higher values of AMI and lower CMI values are observed for MCI-AD patients compared to healthy control both under resting and task conditions. Thus with the advent of signal processing techniques, EEG could be efficiently used as an essential tool for the early diagnosis of the neurological disease-AD and also study the information transmission among different brain regions.