FDA Clears AI Model That Detects Heart Disease From EKG Images

An artificial intelligence model that can identify structural heart disease from standard electrocardiogram images has received Food and Drug Administration clearance, opening the door for widespread clinical deployment through a popular medical search platform.

The technology, called EchoNext, can detect seven distinct forms of structural heart disease by analyzing EKG images. These conditions include problems where heart valves become blocked or leak, preventing proper blood flow, as well as cases where the heart's chambers fail to pump blood effectively.

Researchers at New York-Presbyterian Hospital and Columbia University developed the model, which is now being brought to market by Pathway Labs, a spinout company formed to commercialize the technology. The FDA clearance this month represents one of the most comprehensive approvals for AI-based cardiac screening from electrocardiograms.

Distribution through OpenEvidence

Pathway Labs plans to pursue two parallel commercialization strategies. The company will market EchoNext directly to hospital systems while simultaneously licensing the technology to OpenEvidence, a medical evidence search engine already used by hundreds of thousands of clinicians.

This licensing arrangement marks an unusual distribution approach for medical AI technology. By embedding the tool within OpenEvidence's existing workflow, doctors using the platform will be able to upload EKG images and receive algorithmic predictions about potential structural heart disease without adopting separate software systems.

Why it matters

Structural heart disease often goes undetected in early stages when intervention could be most effective. By making screening accessible through a tool clinicians already use daily, this approach could significantly expand the reach of AI-assisted cardiac diagnosis beyond specialized cardiology departments. The broad FDA clearance for seven different conditions also suggests the model demonstrates reliability across multiple disease presentations, potentially reducing the need for more expensive or time-consuming diagnostic procedures in initial screening.

Clinical workflow integration

The integration into OpenEvidence represents a shift in how medical AI tools reach practitioners. Rather than requiring hospitals to purchase and implement standalone systems, the technology will be available within a search platform that clinicians consult as part of their regular practice. This embedded approach may accelerate adoption by reducing implementation barriers.

Pathway Labs' dual strategy of direct hospital sales and platform licensing suggests the company is betting on multiple pathways to clinical impact. The hospital channel allows for deeper integration into institutional workflows, while the OpenEvidence partnership provides immediate access to a large user base.

These details were first reported by STAT News.