Chapter 3 - Machine learning and deep learning models for early-stage detection of Alzheimer's disease and its proliferation in human brain

Alzheimer's disease (AD) is a neurodegenerative disorder (NDD) with multiple known and unknown causes. Some of the causes include genetic disposition, lifestyle, nonnutritious diet, lack of physical and mental/cognitive exercise, and so on. Just like arteries become clogged due to stress, fatty foods, lack of physical activity, malfunctioning liver, and accumulation of cholesterol plaque in their walls, the brain can become “clogged” due to abnormal build-up of proteins in and around its cells and degeneration of interconnections between neurons, thus resulting in AD. Proteins called amyloids cause plaques in the brain's cells, and the protein called tau causes tangles within neurons. As such, otherwise healthy neurons stop functioning, lose connections with other neurons, and die. Neural network (NN) models like artificial neural networks (ANNs), deep neural network (DNN) models like convolutional neural networks (CNNs), recurrent neural networks (RNNs), fast RNNs, conceptual alignment DNNs, concept bottleneck models, multitask capsule architectural models, and so on have been invented by computer scientists and researchers to help in the diagnosis and prediction of different diseases. This chapter brings together medical and computational domains to discuss the use of deep learning (DL) and machine learning (ML) in the early detection of AD.