Alzheimer’s Disease Classification Using Vision Transformer

The paper proposes a novel approach for enhancing Alzheimer’s Disease (AD) Classification using Magnetic Resonance Imaging (MRI) of the brain from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset. It uses a Convolutional Next v2 Transformer (ConvNext v2) to classify various AD stages. It adopts Gradient-Weighted Class Activation Mapping (Grad-CAM) as an Explainable Artificial Intelligence (XAI) technique, improving the understandability and interpretation of AD stage classification. The accuracy of classifying five different stages of AD was 98.3%. The Grad-CAM visualisations offered essential features verified with clinical MRI features. This way, our proposed approach is a robust, reliable, and interpretable tool for MRI-based image classification of AD.