ALZHEIMER'S DISEASE PREDICTION USING ENHANCED U-NET ARCHITECTURE

Alzheimer's disease (AD) is the most prevalent kind of dementia for which there is now no cure. Correctly classifying AD may facilitate diagnosis and therapy selection. Multiple studies over the last decade have shown that DL algorithms effectively diagnose AD. Based on 3D T1-weighted magnetic resonance imaging, this work presents an E-U-net model for identifying AD (MRI). It has been shown that the Model's performance is boosted when coupled with stringent supervision. Attempts have been attempted to quantify the presence of amyloid-beta in body fluids for clinical diagnosis of Alzheimer's disease. However, these approaches have not been completely successful in predicting cognitive decline in patients. They are not aimed at elucidating the molecular pathogenesis induced by Tau and amyloid-beta proteins in hippocampus neurons of AD patients. This paper proposed Enhanced U-Net architecture for MRI Image segmentation. This study recommends evaluating the segmentation of True and predicted masks by detecting the fluorescence signal inside, which might give valuable insight into the nature of the disease. The experimental results are shown in the input MRI image with corresponding tru