Madhav Donthula

Machine Learning for Alzheimer's Disease

Started July 21, 2023

Abstract or project description

The progression of Alzheimer's disease is not completely understood, and there is still no cure. We plan to train a machine learning model to classify different stages of severity for Alzheimer's disease. Our dataset is from Kaggle, containing a few thousand images but at a low enough resolution to be less than 100 MBs. We plan to train neural networks from scratch using scikit-learn, but we may also experiment with pretrained models to serve as a baseline. As a stretch goal, we may lastly experiment with multi modal models to investigate how to qualitatively describe the difference in severity.