With the aid of artificial intelligence, a common and affordable test available in many medical offices may soon be used to screen for hidden heart disease. Structural heart disease affects millions globally, yet it often goes undetected due to the lack of routine screening tests.
“We have colonoscopies, we have mammograms, but we have no equivalents for most forms of heart disease,” said Pierre Elias, assistant professor at Columbia University Vagelos College of Physicians and Surgeons and medical director for AI at NewYork-Presbyterian.
Elias and his team at Columbia University and NewYork-Presbyterian developed EchoNext, an AI-powered tool that analyzes electrocardiogram (ECG) data to identify patients who should undergo an echocardiogram. A study published in Nature showed that EchoNext accurately identified structural heart disease from ECG readings more frequently than cardiologists.
“EchoNext basically uses the cheaper test to figure out who needs the more expensive ultrasound,” explained Elias. “It detects diseases cardiologists can’t from an ECG.”
The research team ran EchoNext on nearly 85,000 patients undergoing ECGs who had not previously had an echocardiogram. The AI tool identified over 7,500 individuals as high-risk for having undiagnosed structural heart disease. Among these high-risk individuals, 55% went on to have their first echocardiogram, with nearly three-quarters diagnosed with structural heart disease.
“You can’t treat the patient you don’t know about,” Elias stated. “Using our technology, we may be able to turn the estimated 400 million ECGs that will be performed worldwide this year into 400 million chances to screen for structural heart disease.”
The researchers released a deidentified dataset to assist other health systems in improving screening for heart disease and launched a clinical trial across eight emergency departments.
Authors involved in the study include Timothy J. Poterucha and Linyuan Jing from NewYork-Presbyterian among others from institutions like Weill Cornell Medicine and Montreal Heart Institute.
Funding was provided by several organizations including the National Institutes of Health and American Heart Association. Columbia University has submitted a patent application for the EchoNext ECG algorithm.
