Deep Learning based Method for Alzheimer's Disease Stages Classification using MRI Images

Alzheimer's disease, one of the numerous forms of dementia, presents a considerable challenge to medical care systems. Indeed, there is currently no cure, but early diagnosis and prevention of the disease might be the consequence of ineffective treatment. The absence of effective treatments has led many scientists to look for other ways to analyze and detect cases at a premature stage. One of the ways that are receiving considerable interest is the one based on deep learning, which enables computers to learn from massive datasets without requiring human supervision. This has allowed the development of algorithms with high accuracy leading to better results than traditional methods when used with a doctor's medical evaluation. This paper focuses on developing a technique based on a Convolutional Neural Network to classify Alzheimer's disease stages from Magnetic Resonance Imaging data through two distinct scenarios. We compared our results with other state-of-the-art methods, and ours yielded more promising performances.