A Comprehensive Study of Alzheimer's Disease Detection and Its Classification By Deep Learning

Alzheimer's Disease (AD) is a neurogenerative disease, the most familiar type of dementia, and one of the critical brain diseases which are associated with aging. It is the cause of 60-70% of cases of dementia. AD is the sixth major reason of death in the United States of America. Nearly 50 million people suffer from Alzheimer's disease or similar dementia worldwide. The early diagnosis of an AD patient is important because it helps the patient to receive some preventive methods before the appearance of invariable damage in the brain. AD cannot be detected at an early stage, because it can only be detected once it becomes noticeable. Deep Learning (DL) approach in diagnosing AD was more accurate compared to the conventional machine learning models at early stages. This study aims to summarize the most suitable way of predicting and classifying AD by Deep Learning using a comprehensive study. The study also contains a detailed study of various biomarkers and datasets for AD prediction and classification.