Alzheimer’s Disease Detection Using Deep Learning

Abstract

Alzheimer disease (AD) is a neurodegenerative disorder. For the Alzheimer disease, no treatment is specific. Early diagnosis of Alzheimer's disease will allow patients to get proper care. Statistical and machine learning methods are used by most studies to diagnose Alzheimer disease. The human-level performance of Deep Learning algorithms has been successfully demonstrated in various fields. In the suggested method, the MRI information is employed to detect the Alzheimer disease and Deep Learning method is employed to classify the current disease. The classification of Alzheimer's disease based on deep learning techniques has yielded encouraging results, and successful implementation in the clinical environment calls for a blend of high accuracy, minimal processing time, and generalizability to diverse populations. In this research work, we established a system for the detection of Alzheimer's disease based on Convolutional Neural Network (CNN) architecture using magnetic resonance imaging (MRI) scans images which are trained by Kaggle dataset. Models for this research work are trained over the same dataset to compare their performances. The Convolutional Neural Network (CNN) model provides the best accuracy where train accuracy is 86.34% and validation accuracy is 86.45% on the test data that correctly detects Alzheimer disease.