Accuracy-Based Performance Analysis of Alzheimer’s Disease Classification Using Deep Convolution Neural Network

Abstract

Many researchers have utilized various statistical and machine learning models for detection of Alzheimer’s disease. There is been a common practice of analyzing magnetic resonance imaging (MRI) for Alzheimer’s disease diagnosis in clinical research. Based on similarity between healthy and demented MRI data of older people has been done for Alzheimer’s disease detection. Recently, advanced deep learning techniques have successfully illustrated human-level performance in various fields, including medical image analysis, which is advantageous over hand crafted feature extraction methods. Convolutional neural network (CNN) provided better potential for accuracy in diagnosis the Alzheimer’s disease such as to classify the given input as cognitively normal (CN), mild cognitive impairment (MCI) and Alzheimer disease (AD). In this work, we have presented a framework based on DCNN for Alzheimer’s disease detection in terms of accuracy. We have achieved 97.98% accuracy on our dataset without using any handcrafted features for training the network. Validation accuracy achieved is 91.75%. Experimental data is obtained from ADNI and total 13,733 images from 266 subjects are used.