AI-Powered Early Diagnosis and Prognosis Prediction in Neurodegenerative Diseases

Abstract

Progressive loss of neurons is a hallmark of neurodegenerative diseases, including Alzheimer disease (AD), Parkinson disease (PD), and amyotrophic lateral sclerosis (ALS), associated with cognitive and motor dysfunctions 4–8. To enable timely intervention, it is important to diagnose early as well as accurately predict the disease prognosis. AI is the leading innovation that has proven useful for analyzing complex clinical, imaging, and genetic data to facilitate automated, fast, and accurate detection and prediction. We aim to review current AI methodologies for neurodegenerative disease diagnostics and prognosis, perform an empirical analysis of machine learning methods on a public dataset, and discuss future directions. We find that AI algorithms — deep learning and ensemble techniques in particular — predict early detection (91% accuracy) and robust prognosis that outperforms traditional clinical assessments.