On the Detection of Alzheimer’s Disease using Support Vector Machine Based Majority Voter Classifier

Abstract

Early detection of Alzheimer’s disease (AD) has drawn enough attention of researchers throughout the globe because of the lack of well-defined diagnosis of the disease. This has become one of the major threats for the elderly people in particular. This work makes a novel attempt to classify the brain MRI images into two classes viz. AD and non-AD using the volumetric information of white matter (WM), grey matter (GM) and cerebro spinal fluid (CSF). This has been accomplished with the help of three parallel support vector classifiers followed by a majority voter classifier. Performance of this proposition has been measured with the help of accuracy, sensitivity & specificity and subsequently is compared with some of the existing methods.