New AI tool more accurate than cardiologists to spot hidden heart disease
New Delhi, July 17 (IANS/WISHAVWARTA) A newly developed artificial intelligence (AI) tool, that uses data from the low-cost electrocardiogram (ECG), may be more accurate in identifying hidden heart disease than cardiologists, according to US researchers.
Structural heart disease, including valve disease, congenital heart disease, and other issues that impair heart function, affects millions of people worldwide. These often remain undetected due to lack of a routine and affordable screening test.
To address the gap, a team from Columbia University in the US has developed an AI-powered screening tool, EchoNext, that detects structural heart diseases using ordinary ECG data.
EchoNext identifies patients who should have an ultrasound (echocardiogram) — a non-invasive test that is used to diagnose structural heart problems. In the study, published in the journal Nature, the tool was found to be more accurate than cardiologists.
“We have colonoscopies, we have mammograms, but we have no equivalents for most forms of heart disease,” said Pierre Elias, Assistant Professor of Medicine and biomedical informatics at Columbia University Vagelos College of Physicians and Surgeons.
EchoNext was designed to analyse ordinary ECG data to determine when follow-up with cardiac ultrasound is warranted.
“EchoNext basically uses the cheaper test to figure out who needs the more expensive ultrasound,” said Elias.
“It detects diseases cardiologists can’t detect from an ECG. We think that ECG plus AI has the potential to create an entirely new screening paradigm,” the researcher added.
The deep learning model was trained on more than 1.2 million ECG-echocardiogram pairs from 230,000 patients.
In a validation study across four hospital systems, including several NewYork-Presbyterian campuses in the US, the screening tool demonstrated high accuracy in identifying structural heart problems, including heart failure due to cardiomyopathy, valve disease, pulmonary hypertension, and severe thickening of the heart.
In a comparison with 13 cardiologists on 3,200 ECGs, EchoNext accurately identified 77 per cent of structural heart problems. In contrast, cardiologists making a diagnosis with the ECG data had an accuracy of 64 per cent.
“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 and potentially deliver life-saving treatment at the most opportune time,” Elias said.
—
