Johns Hopkins University spinoff Bayesian Health has received 510(k) clearance from the Food and Drug Administration (FDA) for an artificial intelligence tool designed to detect sepsis early. Sepsis, a life-threatening response to infection, is a leading cause of death in U.S. hospitals. The company’s system utilizes electronic health records (EHRs) to analyze patient data, including vital signs, laboratory measurements, medications, and emergency department complaints, to identify potential signs of deterioration. Unlike existing FDA-authorized sepsis tools that require a physician to suspect the condition first, Bayesian’s system provides alerts nearly 2 to 48 hours faster than traditional clinical methods.

The technology was developed by Suchi Saria, a professor at Johns Hopkins University and director of the AI & Healthcare Lab. Saria initiated the research after losing a family member to sepsis in 2017 and has spent more than a decade validating the system’s integration into clinical workflows. The FDA previously granted the technology breakthrough designation in 2023. The tool has already been deployed at several health systems, including the Cleveland Clinic, MemorialCare, and the University of Rochester School of Medicine.

A prospective study published in Nature in 2022 demonstrated the impact of the tool, finding that patients with sepsis whose alerts were confirmed by a clinician within three hours experienced reduced in-hospital mortality rates and lower instances of organ failure. According to a summary of these findings, patients with sepsis were 18% less likely to die in the hospital when clinicians acted on Bayesian’s alerts in time. The company emphasizes that the tool helps clinicians proactively manage patient deterioration rather than reacting to it after symptoms become severe.

Johns Hopkins University spinoff Bayesian Health has received 510(k) clearance from the Food and Drug Administration (FDA) for an artificial intelligence tool designed to detect sepsis early. Sepsis, a life-threatening response to infection, is a leading cause of death in U.S. hospitals. The company’s system utilizes electronic health records (EHRs) to analyze patient data, including vital signs, laboratory measurements, medications, and emergency department complaints, to identify potential signs of deterioration. Unlike existing FDA-authorized sepsis tools that require a physician to suspect the condition first, Bayesian’s system provides alerts nearly 2 to 48 hours faster than traditional clinical methods.

The technology was developed by Suchi Saria, a professor at Johns Hopkins University and director of the AI & Healthcare Lab. Saria initiated the research after losing a family member to sepsis in 2017 and has spent more than a decade validating the system’s integration into clinical workflows. The FDA previously granted the technology breakthrough designation in 2023. The tool has already been deployed at several health systems, including the Cleveland Clinic, MemorialCare, and the University of Rochester School of Medicine.

A prospective study published in Nature in 2022 demonstrated the impact of the tool, finding that patients with sepsis whose alerts were confirmed by a clinician within three hours experienced reduced in-hospital mortality rates and lower instances of organ failure. According to a summary of these findings, patients with sepsis were 18% less likely to die in the hospital when clinicians acted on Bayesian’s alerts in time. The company emphasizes that the tool helps clinicians proactively manage patient deterioration rather than reacting to it after symptoms become severe.