The conventional approach to weight loss, based on the calorie balance model, offers the simple advice, “eat less and move more.” Unfortunately, few people can maintain weight loss over the long term through calorie restriction because the body fights back, with rising hunger and slowing metabolism. An alternative approach to treatment aims to target the underlying driver of weight gain – fat cells overstimulated to hoard too many calories – leading to weight loss with less struggle.
John Brownstein, Ph.D. is an Associate Professor at Harvard Medical School and directs the Computational Epidemiology Group at the Children’s Hospital Informatics Program in Boston. He was trained as an epidemiologist at Yale University. Overall, his research agenda aims to have translation impact on the surveillance, control and prevention of disease. He has been at the forefront of the development and application of public health surveillance including HealthMap.org, an internet-based global infectious disease intelligence system. The system is in use by over a million people a year including the CDC, WHO, DHS, DOD, HHS, and EU, and has been recognized by the National Library of Congress and the Smithsonian. Dr. Brownstein has advised the World Health Organization, Institute of Medicine, the US Department of Health and Human Services, and the White House on real-time public health surveillance.
David Kale from the USC Computer Science Department moderated the session “Machine Learning in Complex Medical Data”. Medical Data exemplifies the “5 V’s” of Big Data: volume, velocity, veracity, variety, and value. The session brought three expert panelists to address how complex medical data can be used to better understand the health care system. These speakers included Drake Pruitt of Lionsolver, Adam Perer from IBM, and Finale Doshi-Velez from Harvard Medical School.
Finale Doshi-Velez is an NSF postdoctoral fellow at the Center for Biomedical Informatics at Harvard Medical School, where her research focuses on developing machine learning techniques to extract patterns in clinical data. She completed her doctoral work at the Massachusetts Institute of Technology. Her research focuses on data-driven approaches to discovering disease subtypes by applying latent variable analyses to clinical data. She also works on predictive models with time-series data.