An efficient deep learning approach for detecting Alzheimer’s disease using brain images

Abstract

Alzheimer’s disease (AD) is a disorder of the brain which causes the loss of memory. This is a successively growing disease which means the severity of it will be upward with the time. In this century, AD is one of the major concerns in the medical arena. The objective of our work is to improve a system that will be able to detect the disease at an early stage. We used efficient deep learning for our project as nowadays, deep learning plays a vital role in every research field. We used MRI of brain for our project using which, our model can identify whether there is sign of Alzheimer’s disease or not. Besides, our model is multiclass classification related which means it can work on different stages of AD.We used different architectures of Convolutional Neural Network (CNN) that are VGG-19, Inception V3 and ResNet- 101 and the accuracy we attained from our models are 88.02%, 89.62% and 91.49% respectively which are pretty decent scores. Since it is an irreversible disorder, we can perceive the significance of detecting the disease at an early stage with good accuracy. Thus, we can state based on the performance of our models that these might play some important role to detect AD.