In parallel with Nightingale Open Science, Ziad Obermeyer and Sendhil Mullainathan also lead Nightingale Open Labs, a research network housed jointly at UC Berkeley and the University of Chicago. We work on research projects that blend machine learning, economics, and medicine, to answer some of the most important questions in health. Here are some of the projects we're working on now:
Drawing on our work showing that machine learning can help doctors diagnose heart attack in the ER, we are designing and deploying a large-scale randomized trial of the algorithm in multiple hospitals across the Providence system. This will teach us about how doctors interact with and adopt algorithms, and whether or not better predictions actually translate into better health outcomes.
In the US alone, 300,000 people drop dead of sudden cardiac death every year. What makes this particularly tragic is that implantable defibrillators could have prevented many of these deaths—if we had just known which patients needed it. Machine learning predictions on risk of sudden cardiac death could one day help target defibrillators to patients who need them.
In ERs across the world, doctors facing bed shortages must decide if patients with respiratory infections like COVID-19 are safe to go home, or need hospital-level monitoring. The current state of medical knowledge is failing here: some patients in the hospital ultimately do not require advanced care, wasting beds; others are sent home, only to deteriorate rapidly. Linking chest x-ray data to pulmonary outcomes will enable us to create tools to optimize triage and diagnosis.