Prediction of alzheimer's disease using data mining techniques: A comprehensive review

Abstract

Acute neurodegenerative disease known as Alzheimer’s Disease (AD) causes memory loss by degenerating brain cells progressively. A deadly cerebrum infection generally influences the old. It directs the decline of biological and cognitive functions of the brain and gradually shrinks the brain, resulting in atrophy. People with AD have a much harder time doing even simple tasks as the disease progresses, the most fundamental tasks, and in the worst case, their brain will stop working at all. Interventions and slowing the progression of AD can only be implemented if the disease is predicted early. In the last decade, number of machine learning (ML) and deep learning (DL) algorithms have been investigated with the intention of developing an automated AD detection system. New horizons in this field have been opened by advances in sophisticated deep learning systems and data augmentation methodologies, and research is moving quickly. Thus, the motivation behind this review is to give an outline of late examination on profound learning models for Alzheimer’s infection determination. We also classify implementation and reproducibility, besides the many data sources, neural network designs, and widely employed evaluation metrics. Our goal is to help interested researchers reproduce and keep up with the latest developments.