You’re reading Part 2 of Paying for AI, a series examining how new clinical artificial intelligence tools influence the affordability of health care and patients’ long-term health. Read Part 1 here.
Five years ago, the bottom fell out of sepsis prediction software. Hundreds of hospitals had adopted an algorithm from electronic health record company Epic that promised to alert physicians to predicted cases of sepsis, a life-threatening reaction to infection that kills more than 350,000 people in the United States every year.
The AI was a technical flop. Despite its results on paper, the technology failed to perform in the real world, and sent so many alerts that doctors tuned them out or hospitals turned them off.
Half a decade on, new sepsis models are hitting the scene. Epic released a retooled version of its own algorithm. Startups are testing their models in health systems. A team uses large language models to mine clinical notes for signs of sepsis. And on Tuesday, a sepsis flagging device from Bayesian Health, with origins at Johns Hopkins, announced it has received clearance from the Food and Drug Administration.
This article is exclusive to STAT+ subscribers
Unlock this article — and get additional analysis of the technologies disrupting health care — by subscribing to STAT+.
Already have an account? Log in
To read the rest of this story subscribe to STAT+.
