The FDA has cleared an artificial intelligence–based early warning system designed to identify sepsis prior to clinical suspicion, according to a press release from Johns Hopkins University. Johns Hopkins described the technology as “one of the first AI-based medical tools to get clearance.”

The Targeted Real-Time Early Warning System, developed by Johns Hopkins researchers and commercialized by Bayesian Health, integrates electronic health record data with clinical artificial intelligence to alert clinicians to possible sepsis. Johns Hopkins reported that the system helped physicians identify sepsis nearly 2 to 48 hours earlier than traditional methods. The press release did not specify how that interval was derived or provide additional methodological details.

The university also reported that the system was associated with an 18% reduction in sepsis mortality rates across dozens of US hospitals. The press release did not specify whether the reported reduction was relative or absolute and did not include study design details, patient population characteristics, comparator definitions, or performance metrics.

“Pre-suspicion screening is what creates lead time, and lead time is what changes outcomes in sepsis,” said Suchi Saria, PhD, a Johns Hopkins professor and director of the AI & Healthcare Lab. “No other cleared test or device monitors for sepsis prior to clinician suspicion.”

In 2023, the FDA granted the technology Breakthrough Designation, which is intended to expedite review of technologies with potential to improve care for life-threatening conditions. Johns Hopkins said the system was subsequently deployed at several health systems, including Cleveland Clinic, MemorialCare in California, and the University of Rochester School of Medicine, where it “significantly reduced in-hospital mortality, morbidity, and length of stays for patients with sepsis.” The press release did not provide supporting data for those findings.

Sepsis can be difficult to identify because symptoms such as fever and confusion overlap with those of other conditions. Johns Hopkins noted that delayed detection is associated with lower survival, underscoring the rationale for earlier warning systems.

According to Johns Hopkins, FDA clearance may allow hospitals using the system to seek Medicare and Medicaid reimbursement through the New Technology Add-on Payment program.

Source: Johns Hopkins Press Release