A review of deep learning-based disease detection in Alzheimer's patients

Abstract Early and accurate diagnosis is a significant step in lowering the risk of Alzheimer's progression. Brain-imaging techniques can help pinpoint the abnormalities related to Alzheimer's. Using artificial intelligence-driven systems, brain images are classified into healthy and abnormal based on the extracted features. These features are capable of precisely detecting the Alzheimer's-related symptoms by identifying changes in the brain segments such as ventricle size, hippocampus shape, cortical thickness, and brain volume. This chapter discusses how Alzheimer's affects different parts of the brain, the stages of Alzheimer's, neuroimaging techniques that are being used to visualize the effects, resources of relevant datasets, and how different AI-based methods developed over the years are able to identify changes in the brain and then formulate the defining features. Finally, we compare all the discussed methods with a relevant performance metric.