Abstract
Alzheimer's Disease (AD) is normally identified by several behavioral symptoms often mistakenly associated to age-related concerns or stress. However correct diagnosis and monitoring of the disease requires of additional resources. This paper presents a new methodology for classification of Alzheimer's disease from MR images for medical support. A large database with more than one thousand patients was used. Two different problems are tackled in this work: a first one where a classification method is developed to classify MR images as either normal or with the Alzheimer's disease and a second one for the identification and classification between normal subjects, MCI patients and AD patients. It is noteworthy that with this last study we could offer a tool to assist the early diagnosis of dementia. The outline of the methodology includes wavelet feature extraction from the MRIs, dimensionality reduction, training-test subdivision and classification using Support Vector Machines. Some concerns related to performance evaluation and dimensionality reduction are discussed.

This publication has 14 references indexed in Scilit: