Alzheimer’s Detection using Machine Learning and Deep Learning: A Literature Survey

Abstract—This literature survey investigates recent advance- ments in Alzheimer’s disease (AD) detection through the lens of machine learning and deep learning methodologies. The survey explores a variety of research papers, each offering unique perspectives on early diagnosis and classification. Deep learn- ing architectures, particularly Convolutional Neural Networks (CNNs), emerge as highly successful in AD detection. Innovative approaches include the use of smartphone accelerometers for disease stage prediction, displacement field estimation as AD- related features, and the integration of machine learning with electronic health records for practical clinical applications. While these advancements showcase significant progress, challenges such as dataset dependency, model interpretability, and ethical considerations persist. Future research directions point towards multimodal data integration, enhancing model interpretability, and addressing ethical aspects for more robust and transparent AD detection models. Index Terms—Alzheimer’s, machine learning, deep learning, CNN, early detection, multimodal data, interpretability, ethical considerations