Detecting different stages of Alzheimer’s disease from MRI images using deep learning and computer vision techniques

Abstract

The preliminary and precise diagnosis of Alzheimer’s Disease is significant for the speedy management and intervention of the disorder. Numerous valuable tools such as Magnetic Resonance Imaging (MRI), Positron Emission Tomography (PET) etc. are used for evaluating the function and structure of brain which could help diagnose Alzheimer’s Disease. However, simplifying the MRI images manually is a hectic and long drawn process that is prone to observer variability. The prime prospect of this study is to employ a Computer Vision and Deep learning Based framework that would automatically classify the stage Alzheimer Disease (AD) through the MRI images. A large amount of dataset is used to enhance the effectiveness of the suggested structure. Moreover, this research demonstrates the capability of Computer vision and Deep learning in assisting premature AD detection. It provides a beneficial insight into the enrichment of neurological disease diagnosis using computer-aided technology. The highlight of this study is the introduction of a custom model that outperforms all state-of-the-art Convolutional Neural Network (CNN) models in performance. This novel model has achieved an exceptional accuracy of 96.6%, which showcases a meaningful advancement in the field and also provides a promising direction for future research in neurodegenerative disease diagnosis.