Identification of Alzheimer’s Disease using Functional Connectivity Measures

Abstract

Alzheimer’s disease (AD) is a gradual neuro cognitive disorder caused by the damage of brain cells over a certain period of time. One non-invasive and efficient technique to investigate AD is to use functional magnetic resonance imaging (fMRI). Functional connectivity is a change in the functional connections between brain regions when an activity takes place. The correlation value gives the strength of functional connectivity. Pearson’s correlation method was used to calculate the correlation coefficient between two time series. Mutual information which denotes the information successfully transmitted through a channel was also considered. In this paper, these two measures are compared and their performance and suitability is assessed for fMRI connectivity modelling based on the classification accuracy. Machine learning techniques such as support vector machine (SVM) is employed for connectivity analysis and classification of Alzheimer’s from control population.