EchoNext
Pathway Labs · FDA clearance
AI flags hidden structural heart disease from a routine 12-lead ECG.
A neural network, developed at NewYork-Presbyterian and Columbia and trained on more than 700,000 paired ECG and echocardiogram studies, that flags patients at high risk of structural heart disease who should have an echocardiogram. On 3,200 ECGs it identified 77% of structural heart problems, against 64% for cardiologists reading the same tracings. It is a screening aid only; echocardiography is still required for diagnosis.