The Food and Drug Administration (FDA) has cleared an artificial intelligence (AI)-based sepsis detection system for approval.
The Targeted Real-Time Early Warning System, developed by researchers at Johns Hopkins University and commercialized by Bayesian Health, integrates electronic health records with advanced clinical AI to continuously monitor patients and flag sepsis up to 48 hours before a clinician suspects it. A 2022 study of more than 764,000 patient encounters at five US hospitals found that when clinicians acted on the tool’s alerts, sepsis patients were 18% less likely to die in the hospital.
Sepsis occurs when the body’s immune system has an overwhelming reaction to an infection, causing a rapid chain of events leading to tissue damage, organ failure, and death. At least 1.7 million US adults and more than 18,000 US children develop sepsis each year, and at least 350,000 adults and more than 1,800 children who develop the condition die during their hospitalization.
A ‘needle-in-a-haystack problem’
Effective treatment and better patient outcomes for sepsis rely on catching it early, which can be difficult because sepsis symptoms, including fever and increased heart rate, are common in other medical conditions. Each hour of delayed treatment can reduce survival by 8%.
Company officials say the platform is the first FDA-cleared, AI-based device that detects sepsis before clinical suspicion.
“Catching sepsis before a clinician suspects it is a needle-in-a-haystack problem,” Neri Cohen, MD, PhD, head of clinical enterprise at Bayesian Health, said in a company press release. “Missing a single case is catastrophic, and that demands a level of precision most AI can't meet, even tools that look promising in preliminary studies."
The receipt of FDA 510(k) clearance means the agency has deemed the device is “substantially equivalent” to a device that’s already on the market.