Abstract—According to world Health Organization (WHO) India’s population of those aged over 65, which was 40 million in 1997, is to increase 108 million by 2025 and 240 million by 2050. This means several fold increase in age related problems such as neuro-degeneration, a condition characterized by a progressively declining memory and intellectual functions. In this paper, we propose an Artificial Neural Network based method to evaluate Dementia severity in terms of CDR scale and in turn to classify various Dementia levels. We show that a Multilayer perceptron (MLP) neural network can classify Dementia levels to the accuracy of about 95%, after a process of supervised training. The study has been performed on dataset of 416 cases, available at the Open Access Series of Imaging Studies (OASIS) database, which is available at http://www.oasis-brains.org.
Index Terms—Neuro-degeneration, dementia, CDR (Clinical Dementia Rating), artificial neural network, MMSE (Minimental Status Examination), multilayer feed forward network, OASIS data set.
M. M. Patil and A. R. Yardi are with the SVERI’s College of Engg. Pandharpur (e-mail: mns.patil@gmail.com; aryardi@yahoo.com).
T. C. Author is with the Electrical Engineering Department, University of Colorado, Boulder, CO 80309 USA, on leave from the National Research Institute for Metals, Tsukuba, Japan (e-mail: author@nrim.go.jp).
Cite: M. M. Patil and A. R. Yardi, "MLP Classifier for Dementia Levels," International Journal of Modeling and Optimization vol. 2, no. 5, pp. 567-569, 2012.
Copyright © 2008-2024. International Journal of Modeling and Optimization. All rights reserved.
E-mail: ijmo@iacsitp.com