Charlottesville, Va.-based UVA Health integrated an artificial intelligence model developed by one of its physicians into its remote monitoring technology to predict patients' risk of a serious event over the next 12 hours.
CoMET, the software designed by UVA Health cardiologist Randall Moorman, MD, combines remote monitoring with AI algorithms to predict patients' risk of a serious event so that medical staff can stabilize their vital signs before adverse events occur, such as sepsis, blood poisoning, respiratory distress or cardiac instability.
The model updates every 15 minutes and uses numbers that are drawn from a patient’s EKG every two seconds. It displays patients' risk on a large LCD screen, showing small comet-like figures that hover near the X-Y axis for stable patients. As a patient's risk level rises, the comets grow, turn orange or deep red and move up the screen like shooting stars.
"Vital sign measurements and labs can come too late, but early detection through predictive analytics has the power to improve patients’ outcomes, especially for catastrophic illnesses like COVID-19," Dr. Moorman said in a Jan. 21 news release.