Every hour treatment for sepsis is delayed, the risk of the patient dying increases by as much as 8%, making the systemic infection a leading cause of death in hospitals. But sepsis is hard to diagnose because its symptoms (such as fever) are so common in hospital patients. The Targeted Real-Time Early Warning System–developed by researchers at Johns Hopkins University and Bayesian Health–uses AI to analyze a patient’s medical records, symptoms, and lab results, alerting a clinician to probable sepsis cases. “The typical early-alert systems run at about 20% to 30% sensitivity,” says Suchi Saria, founder and CEO of Bayesian Health. “Our system was able to catch 82%.”
More Must-Reads From TIME
- What a Photographer Saw in the West Bank
- The Dirty Secrets of Alternative Plastics
- Accenture’s Chief AI Officer on Why This Is a Defining Moment
- We Should Get Paid for Our Online Data: Column
- Inside COP28's Big 'Experiment'
- The 100 Must-Read Books of 2023
- The Top 100 Photos of 2023
- Want Weekly Recs on What to Watch, Read, and More? Sign Up for Worth Your Time
