A Deep Learning Approach for Early Diagnosis of Alzheimer’s Disease using MRI Images

Abstract

This dissertation focuses on Alzheimer's disease diagnosis using deep learning techniques applied to MRI images. With the rapid advancements and changes in technology and AI techniques, they could aid in diagnosis and classification while posing no significant risks by utilizing existing MRI image data. Although there are no working treatments yet, only medicines that can help slow the progression of the disease, early detection and classification could help choose the best treatment plan. We chose the transfer learning method, employing two well-known pre-trained CNN models (VGG16 & MobileNetV2). The experimental results show that our proposed approach outperforms other approaches and methods in terms of accuracy, achieving 99.71% with VGG16 and 100.0% with MobileNetV2.