abstract
Alzheimer’s Disease (AD) is a progressive neurodegenerative condition that affects memory, cognition, behavior, and other cognitive functions. It is regarded as a silent global health crisis due to alarming statistics and projections for the future. However, there are currently no effective treatments or precise diagnostic methods available. Despite the absence of a cure, early detection of AD is paramount for developing therapeutic and care plans that aim to preserve cognitive function and prevent irreversible decline. In this Systematic Literature Review (SLR) we delve into the potential contribution of AI to aid in the detection of Alzheimer’s Disease. Firstly, we outline the methods presently employed for diagnosing AD, which include both invasive and non-invasive biomarkers. Focusing exclusively on Magnetic Resonance Imaging (MRI) as the primary neuroimaging technique, we underscore the significance of data preprocessing. Subsequently, we investigate the potential contributions of AI and outline three fundamental approaches: supervised learning, Weakly Supervised Learning (WSL), and transfer learning, the latter arising as a bridge between supervised learning and WSL/unsupervised learning. Finally, the paper addresses the most relevant legal and ethical implications, alongside the necessary skills enhancement required for healthcare professionals to effectively implement AI in the ethical medical domain. A description of a possible scenario is also provided as a suggestion to improve the quality of the healthcare system.
outcomes
