A Comprehensive study on Accurate Brain Disorder Prediction using ML-DL frameworks for Alzheimer’s Disease Analysis

Alzheimer's Disease (AD) is one type of progressive neurodegenerative disease that negatively affects the cognitive ability of patients, requiring an early detection technique to properly intervene. Conventional diagnostic techniques have limitations in terms of costs, intrusiveness, and their incapability of detecting early-stage changes. Over the past few decades, Artificial Intelligence (AI) especially Machine Learning (ML) and Deep Learning (DL) has gained popularity due to its success in predicting different diseases. ML and DL techniques alone are facing issues regarding feature representations, generalization of knowledge, and interpretability. This study proposes a thorough analysis of hybrid models consisting of ML and DL techniques for predicting early-stage Alzheimer's Disease based on various types of modalities including MRI, PET, EEG, and clinical records. Initially, a thorough literature review will be conducted followed by the determination of major gaps in the current state of research such as data availability, model generalization, and computational issues.