University of Chicago Medicine and Google will join forces to investigate how historical patient data stored in EHRs can improve future health outcomes.
The UChicago Medicine project will leverage machine learning to predict hospital readmission. The collaboration will unite researchers with predictive modeling expertise from UChicago Medicine's Center for Healthcare Delivery Science and Innovation and its Center for Research Informatics with machine-learning tools developed by Google researchers.
"Traditional tools of epidemiology and statistics simply can't use free text or images to create predictive algorithms that could alert physicians and nurses about patients' risks for problems," said Michael Howell, MD, director of the Center for Healthcare Delivery Science and Innovation. "But together with Google, we can."
UChicago Medicine is the latest institution to join Google's research on predictive models in healthcare, following Stanford (Calif.) University and UC San Francisco.