Data Science Day @ Columbia University

As more and more observations are recorded in patient records, providers face an overwhelming amount of complex data points, with little time for making sense of them all. This phenomenon of information overload has been observed in primary and specialty care, during hospital admissions, and in the emergency room. One of the promises of the electronic health record is to support clinicians at the point of care. Unfortunately, it seldom provides effective cognitive support. HARVEST is an interactive, problem-oriented patient record summarization system. It innovates in three ways: (i) it extracts content from the patient notes, where key clinical information resides; (ii) it aggregates and presents information from multiple care settings, including inpatient, ambulatory, and emergency department encounters; and (iii) it is integrated into the electronic health record at NewYork-Presbyterian Hospital.