FDA Clears AI Tool That Detects Hidden Heart Disease From a Standard ECG

The FDA has cleared EchoNext, an artificial intelligence tool that can detect structural heart disease from a standard 12-lead electrocardiogram, Pathway Labs announced in June 2026.

The clearance covers six structural heart conditions: right-sided heart failure, left-sided heart failure, valve disease, severe thickening of the heart muscle consistent with infiltrative cardiomyopathy, and pulmonary hypertension.

EchoNext was developed by researchers at NewYork-Presbyterian and Columbia University. The tool analyzes ECG waveforms to flag patients who may need further evaluation with an echocardiogram, a more detailed imaging test.

In a 2025 validation study published in Nature, EchoNext correctly identified 77% of structural heart problems. Cardiologists reviewing the same ECG data achieved 64% accuracy. The study found no clinically relevant differences in the tool''s accuracy across racial, ethnic, or sex-based demographic groups.

A peer-reviewed case in Nature Medicine documented the tool identifying severe, undiagnosed heart failure in a patient, leading to timely intervention and a heart transplant.

Pathway Labs has partnered with OpenEvidence, a medical search platform used by hundreds of thousands of clinicians, to make EchoNext available at the point of care. The company recently raised $8.5 million in seed funding from AlleyCorp and Breyer Capital to expand the tool across additional health systems.

Medical experts say AI tools like EchoNext should support clinical judgment, not replace it. Cardiologists stress the importance of taking detailed patient histories and remaining alert to potential biases in training data, even when AI results look strong.