Classifying Alzheimer's disease and tracking disease progression using machine learning methods Lead Investigator: Yueqi Ren Institution : University of California, Irvine E-Mail : yueqir@uci.edu Proposal ID : 1634 Proposal Description: 1. Use supervised machine learning (ML) methods to classify cognitively-normal vs AD subjects based on cognitive, MRI, PET, and CSF biomarkers serves to both test various ML pipelines and validate published work in the face of existing difficulties to replicate ML results. 2. Apply deep learning methods to cluster subjects based on cognitive, MRI, PET, and CSF biomarkers serves to identify sub-clusters within the patients group and compare with results from supervised ML methods. 3. Using subjects within both the Uniform Data Set and Neuropathology Data Set, track progression over time to build a ML model for disease progression of patients with confirmed AD diagnosis post-mortem.