International Journal Of Imaging

Abstract

SENTIMENT BASED DRUG RECOMMENDATION USING MACHINE LEARNING

Author: N.Lavanya, Rosline, M.Rohith Reddy, M. Santhosh naik, N. Srikanth,

Abstract:

Early detection of Alzheimer's disease (AD) is based on the categorization of characteristics retrieved from brain scans, which plays a significant role in preventing and treating the illness. The characteristics must properly reflect key AD-related variations in physical brain structures such ventricles size, hippocampus shape, cortical thickness, and brain volume. This research suggests using a deep 3D convolutional neural network (3D-CNN) to predict Alzheimer's disease using generic features that capture AD biomarkers and can adapt to different domain datasets. The 3D-CNN is based on a pre- trained 3D convolutional autoencoder for capturing anatomical shape changes in structural brain MRI data. For each task-specific AD classification, the fully linked higher layers of the 3D-CNN are fine-tuned. ADNI MRI experiments