Suchi Saria, an assistant professor of computer science in the Whiting School and of health policy in the Bloomberg School, leads the seminar. Titled "Individualized Prognosis of Disease Trajectories: Application of Scleroderma," it is scheduled for 11 a.m. on Tuesday, Sept. 8, in Levering Hall's Sherwood Room.
Abstract: For many complex diseases, there is a wide variety of ways in which an individual can manifest the disease. The challenge of personalized medicine is to develop tools that can accurately predict the trajectory of an individual's disease. Access to such tools can help clinicians tailor therapy to the individual. We propose a hierarchical latent variable model that shares statistical strength across observations at different resolutions: the population, subpopulation, and individual level. We describe an algorithm for learning population and subpopulation parameters offline, and an online procedure for dynamically learning individual-specific parameters. We validate our model on the task of predicting the course of interstitial lung disease, one of the leading causes of death among patients with the autoimmune disease scleroderma. We demonstrate significant improvements in predictive accuracy over state of the art.
This is joint work with Peter Schulam (PhD student), Colin Ligon (clinical fellow), and Fredrick Wigley and Laura Hummers at the Hopkins Scleroderoma Center.