Fuzzy classification of Alzheimer's disease using statistical moments

Alzheimer's disease (AD) is a neurodegenerative disease of the central nervous system. Automated system for classification between AD and Normal control patients was constructed in this paper using fuzzy logic approach. Statistical central moments of magnetic resonance imaging (MRI) voxel intensities over 24 the most discriminant regions of interests (ROIs) are used as input features for fuzzy inference system (FIS). Area under the curve (AUC) was estimated using k-fold cross-validation methodology as the characteristics of classification performance. It is defined, that simultaneous employing first, second and third statistical moment in FIS results in highest AUC=0.895.