A comprehensive review on early detection of Alzheimer's disease using various deep learning techniques

Alzheimer's disease (AD) is a type of brain disease that makes it hard for someone to perform daily tasks. Early diagnosis and classification of the condition are thought to be essential study areas due to the speedy progression of the disease in people living with dementia and the absence of precise diagnostic procedures. One of the main aims of the researchers is to correctly identify the early stages of AD so that the disease can be prevented or significantly reduced. The main objective of the current review is to thoroughly examine the most recent work on early AD detection and classification using the deep learning (DL) approach. This paper examined the purpose of an early diagnosis of AD, the various neuroimaging modalities, the pre-processing methods that were employed, the maintenance of data, the deep learning used in classifying AD from magnetic resonance imaging (MRI) images, the publicly available datasets, and the data that were fed into the deep models. A comparative analysis of different classification methods using DL techniques is performed. Further, the paper discussed the challenges involved in AD detection.