Applying AI to bring better outcomes to critically ill patients
Description: An AI tool that enables the physician early detection of life-threatening events and assists in daily decisions. The researchers developed an ML algorithm that analyzes the available data in the patients’ EMR in order to recognize blood-culture samples with a high probability to be positive for bloodstream infection at the sampling time (24 hours before it is feasible today).
Status: The algorithm reached excellent PoC at Rambam Health Care Campus and Beth-Israel Deaconess Medical Center, Boston, Massachusetts. Preliminary results for early detection of other events show good performance. Patent pending – US.