Loading...
New articles about artificial intelligence–related decision-support tools are being published at a rapid pace. These investigators assessed the accuracy of an algorithm to predict left ventricular systolic dysfunction (LVSD) among people presenting with dyspnea to the emergency department (ED). The premise was that the rapid identification of LVSD, especially in settings where ultrasound is not readily available, could facilitate clinical decision making.
The researchers retrospectively identified 1606 people with dyspnea through claims codes (median age, 68; 47% women; 91% white) and applied the algorithm to the first available electrocardiogram. The outcome with the algorithm was compared with the result from an echocardiogram performed wi…