Innovative Approach for Alzheimer's Disease Detection

Abstract

Alzheimer's disease (AD) presents a pressing global health challenge, with its prevalence expected to rise in the coming years. Early detection and management becomes crucial to improve the health of the patient. However, current diagnostic methods face limitations, including the reliance on manual interpretation of MRI images and the lack of comprehensive tools for disease management. To address these challenges, this research paper proposes a holistic approach to AD detection and management. Leveraging deep learning techniques, particularly transfer learning. Building upon existing literature the research study focuses on the key challenges faced during detection of the disease. This is followed by developing the Alzheimer's detection system using transfer learning approach, observing the output and its performance impact using AUC approach, respective learning rate along with the batch size that has been considered. The algorithm is further modified using hyper tuning parameters and data augmentation . The best algorithm between inceptionv3 with additional capacity and modified inceptionv3 is determined. The paper concludes by paving a path for utilizing different state of art technologies for Alzheimer's detection.