On June 23, 2026, NewYork-Presbyterian and Columbia University Irving Medical Center announced that EchoNext, an AI tool developed by their researchers, had received what they call the world’s first FDA clearance for AI detection of hidden structural heart disease from electrocardiograms. The model reads standard 12-lead ECGs to flag conditions such as valve disease, cardiomyopathy, pulmonary hypertension, and congenital heart disease that clinicians often cannot see on an ECG alone.
EchoNext was trained on more than 700,000 ECG-echocardiogram pairs and validated across four hospital systems. In a head-to-head comparison it achieved 77 percent accuracy in identifying structural heart problems, versus 64 percent for 13 cardiologists. Dr. Pierre Elias, medical director for AI at NewYork-Presbyterian, said the tool can diagnose critical, high-risk conditions that the human eye cannot.
Elias founded Pathway Labs to commercialize EchoNext. The company separately announced an 8.5 million dollar seed round and a partnership with OpenEvidence to distribute the tool, and a related case was published in Nature Medicine on June 22, 2026.