Alzheimer Disease Detection Using MRI: Deep Learning Review

Deep learning for Alzheimer disease detection using MRI is an emerging area of research in medical image processing. With the advent of new technologies based on methods of Deep Learning, medical diagnosis of certain diseases has become possible. Alzheimer’s is a disease which till date has no cure but the progression of the disease can be slowed down or a person who might develop Alzheimer in future can be treated if early prediction of it is possible. Early prediction of the disease benefits medical professionals a lot for the correct diagnosis. Medical professionals label Alzheimer patients based on the progression of the disease as AD (Alzheimer’s), CN (cognitive impairment) and MCI (mild cognitive impairment). In literature many Deep Learning models are used for the early detection of Alzheimer’s disease. Though there are many image modalities, MRI images being non-invasive are considered best for these types of medical experiments. In the present study, we have studied the evolution of Alzheimer’s disease over time, research gaps, challenges towards building advanced models, possible recommendations to overcome those challenges and determining the best performance model. We have focussed on an exhaustive and comprehensive survey of very deep learning-based research papers on Alzheimer’s disease detection. The present work will benefit researchers by providing a clear direction for future scope in Alzheimer disease detection and analysis.