WORCESTER—UMass Memorial Health has expanded an artificial intelligence–powered clinical decision tool to emergency departments across its network, aiming to support frontline nurses with early sepsis detection and improved patient triage.
The KATE AI platform, which was initially piloted at UMass Memorial Medical Center’s University and Memorial campuses in 2023, is now live at five additional sites: HealthAlliance – Clinton Hospital’s Clinton and Leominster campuses, Harrington Hospital’s Southbridge and Webster campuses, and Marlborough Hospital.
Developed by California-based Mednition, KATE AI integrates with the hospital system’s Epic electronic health record system and is intended to deliver real-time insights to nurses assessing incoming patients.
“We are thrilled that KATE AI will now be utilized across the UMass Memorial Health system,” said Ken Shanahan, senior director of emergency medicine and behavioral health, in a statement released by the hospital. “We’ve seen Kate AI streamline workflows for our caregivers.”
The platform’s sepsis detection model is one of its most significant features. According to data released by Mednition earlier this year, the model achieved an Area Under the Curve (AUC) of 99% in identifying sepsis cases based on the Sepsis-3 criteria. The results, based on a retrospective cohort of more than half a million patients across 16 hospitals, also showed a 95% sensitivity and 96% specificity—metrics that signal strong potential for earlier, more accurate diagnosis in emergency settings.
Mednition said the model builds on a prior version already in use during ED triage, which achieved an AUC of 94% even before lab results were available. The company noted that many AI tools in healthcare have struggled to balance accuracy with transparency, and stated its intent to publish full model performance metrics to establish new standards in clinical AI evaluation.
The triage-focused system also aims to reduce the number of patients who leave without being seen by flagging high-acuity cases faster and helping staff manage lower-priority cases more efficiently. The tool is designed to reduce “alert fatigue” and bolster nurse confidence by surfacing only relevant, actionable information.
UMass Memorial Health, Central Massachusetts’ largest healthcare system, is the clinical partner of UMass Chan Medical School and serves more than 20,000 caregivers across multiple hospitals and care sites.
Mednition’s KATE AI platform recently received FDA Breakthrough Device Designation for its triage-based sepsis screening model and was recognized as “Best in Show” at the 2025 HIMSS Global Health Conference for its application in high-demand hospital settings.
Have news, tips, or a story worth telling? Reach Editor Charlene Arsenault at carsenault@theworcesterguardian.org—because good stories (and great scoops) deserve to be shared.
