Identification of early-stage Alzheimer׳s disease using SFAM neural network

Alzheimer׳s disease (AD), the most common form of dementia, is a complex and very serious nervous disease. Currently no medication is really effective against it. Even, its diagnosis and management remain challenging research problem for scientists. This paper aims towards identifying early-stage AD based upon the characteristics of the non-oscillatory independent components (ICs) of the auditory event related potential (AERP) waveforms of an oddball task for healthy and newly diagnosed AD subjects. Using 27 sensors to record P300 Evoked Potentials (EPs) as features vectors to train the simplified fuzzy adaptive resonance theory map (SFAM) neural network as a classifier, normal and AD subjects were classified with higher than 95% of success. The use of the ARTMAP-familiarity discrimination (ARTMAP-FD) shows that the separation of the two populations was achieved with high success and without any mistake.