The FDA has cleared an artificial intelligence program designed to predict which patients in an intensive care unit will begin to deteriorate, giving clinicians up to eight hours of advanced warning and time to prepare.
Developed by CLEW, the system gathers data from electronic health records, vital sign monitors and connected ICU devices, and uses machine learning models to calculate each person’s risk of their heart and blood flow becoming unstable.
The full clearance follows up on an emergency authorization granted by the agency last year, aimed at the COVID-19 pandemic, with a focus on spotting the early signs of respiratory distress.
The CLEWICU system also identifies patients at a low risk of decompensating in near real-time, allowing staff to better prioritize the use of their resources.
"We are proud to have received this landmark FDA clearance and deliver a first-of-its-kind product for the industry, giving healthcare providers the critical data that they need to prevent life-threatening situations," CLEW CEO Gal Salomon said in a statement.
Previous studies by CLEW showed the system was able to provide accurate alerts to staff by a median of three-and-a-half hours early. The company also plans to develop AI models predicting patient deterioration across all care settings.