Health Affairs September 16, 2020
Predicting prognosis is as old as medicine itself. Recent breakthroughs in artificial intelligence (AI) and machine learning have led to algorithms that promise to answer one of life’s ultimate questions: When will I die? Analyzing electronic health record (EHR) and other data, different algorithms now exist to predict mortality, seemingly with unprecedented accuracy and without the direct input of clinicians or patients.
The ongoing COVID-19 pandemic has arguably further stoked the fire of this desire to know patient prognoses. Accurately predicting mortality due to COVID-19 would supposedly offer a technological way out of vexing resource allocation decisions. Patients could be ranked by likelihood of survival, and scarce resources (such as ventilators) could be allocated accordingly to maximize lives saved....
Today's Sponsors
Today's Sponsor