Instinctive classification of Alzheimer´s disease using FMRI, pet and SPECT images

Alzheimer´s disease (AD) is the most common form of dementia which mostly affects the aged people. In this Dementia is brain disorder that seriously distresses a person´s ability to carry out day by day behavior. The diagnosis of the Alzheimer´s disease (AD) based on nonnegative matrix factorization (NMF) and support vector machines (SVM) using to identify the disease. The different brain image databases are selected: functional magnetic resonance image (fMRI), single photon emission computed tomography (SPECT) database and positron emission tomography (PET) images. The images collected from the above methods containing for both Alzheimer´s disease (AD) and well controls (normal active brain) patient´s images. These resulting images are applying for the Fisher discriminant ratio (FDR) and nonnegative matrix factorization (NMF) which is used for characteristic selection and Extraction. The finally resulting NMF-transformed sets of data, used to classify by means of a SVM-based classifier.