Applying deep learning to MRI and clinical data for precision medicine Lead Investigator: Xiaotong Jiang Institution : University of North Carolina at Chapel Hill E-Mail : xiaotong.phoebe.jiang@gmail.com Proposal ID : 1369 Proposal Description: The goal of the proposed research is to find the optimal personalized plan for modifiable risk factors by applying deep learning techniques to MRI imaging and patient clinical data in the NACC data. Our bodies react to treatments differently and patient differences often make the clinical decision making hard. When is the best time to treat what patient with what intervention? Precision medicine is a rising area that tries to solve this kind of problem but most of its work has been focusing on low-dimensional data. We want to borrow the strength and flexibility of deep learning and make use of imaging data, together with standard health record data, to help clinicians make better treatment recommendations. We believe this research will open doors to lots of other opportunities for precision medicine to take advantage of big, complex data. We have read through the documentation and handbook provided on this website and found the UDS dataset as well as the MRI imaging data good sources of our clinical application.