Advancements in Alzheimer's disease classification using deep learning frameworks for multimodal neuroimaging: A comprehensive review

Abstract

Over the past years, Alzheimer's disease has emerged as a serious concern for people's health. Researchers are facing challenges in effectively categorizing and diagnosing the different stages of Alzheimer's disease (AD). Current promising studies have shown that multimodal Neuroimaging has the potential to offer vital information about the structural and functional alterations associated with Alzheimer's. Using advanced computational techniques, Machine Learning calculations have been demonstrated to be highly precise in deciphering patterns and connections within the multimodal Neuroimaging data, eventually aiding in the arrangement of Alzheimer's illness stages. This research aimed to survey the adequacy of Machine Learning techniques in correctly categorizing stages of Alzheimer's disease by working on multiple neuroimaging modalities. In this review, a detailed analysis was carried out on the classification algorithms included. The study specifically examines publications published between 2016 and 2024. From the review, it was found that deep learning frameworks are more robust in Alzheimer's disease classification.