Predecting the Stages of Dementia Using OASIS

Authors- Vempalli Fazil, Kaluva Bhumika Reddy, Chiranjeevi Althi, Vankadari Lakshmi Bhargavi

Abstract-– Dementia is a debilitating neurodegenerative disorder characterized by progressive cognitive decline, memory impairment, and behavioral changes. Alzheimer’s disease (AD), the most common form of dementia, accounts for 60- 80% of cases worldwide. Early and accurate diagnosis is crucial for effective intervention, yet traditional diagnostic methods often fail to detect early-stage dementia due to their reliance on subjective clinical assessments. This study leverages the Open Access Series of Imaging Studies (OASIS) dataset, a comprehensive repository of neuroimaging data, demographic information, and clinical assessments, to develop a machine learning (ML)-based predictive model for dementia staging. We employ a multi- algorithm approach, including Support Vector Machines (SVM), Random Forest (RF), and Deep Learning architectures (CNN, MobileNet, ResNet), to classify individuals into distinct stages: Normal Cognitive Function Mild Cognitive Impairment (MCI) Early-Stage Alzheimer’s Disease Advanced Alzheimer’s Disease Our results demonstrate that Random Forest achieves 92% accuracy, while CNN and ResNet models excel in detecting subtle neuroanatomical changes. This research underscores the potential of AI-driven diagnostics in revolutionizing dementia care.

DOI: /10.61463/ijset.vol.13.issue2.326