Classifying Alzheimer's Disease Stages Using Convolutional Neural Networks on the OASIS MRI Dataset

Alzheimer's Disease (AD) is a progressive neurodegenerative disorder., where early detection is crucial for improving patient outcomes and slowing disease progression. This study presents a Convolutional Neural Network (CNN)-based model to classify AD stages: Mild Dementia., Moderate Dementia., Very Mild Dementia., and Non-Dementia using the OASIS MRI dataset. The dataset comprises 86.,437 MRI scans., with significant class imbalance addressed through weighted sampling and data augmentation. Training (60.,506 images), validation (12.,965 images), and testing (12.,967 images) subsets were used to ensure balanced learning and robust evaluation. The model effectively captured structural patterns unique to each AD stage., achieving a high test accuracy of 98.80%, along with excellent precision., recall., and F1 scores across all classes. These findings highlight the potential of deep learning as a reliable tool for early detection and precise staging of Alzheimer's Disease., especially in distinguishing challenging cases like Non-Demented and Moderate Dementia.