PH affects up to about 1% of the global population, and early detection could reduce morbidity and mortality by enabling timely intervention.

We’ve reached a milestone: Anumana’s ECG-AI pulmonary hypertension algorithm has received FDA 510(k) clearance, becoming the first PH tool cleared for standard 12-lead ECGs and enabling earlier detection within routine clinical workflows.

This clearance is part of Anumana’s broader ECG-AI portfolio, including LEF and other cardiovascular tools, with plans to expand AI-enabled insights across cardiology.

Mayo Clinic co-founded Anumana, and the company emphasizes that the clearance broadens access to AI-enabled insights at the point of care and integrates into everyday ECG-based workflows.

Pulmonary hypertension is a progressive, potentially life-threatening vascular disease that affects the lungs and right heart, with diagnosis often delayed by years; earlier detection is clinically valuable.

The technology runs within existing clinical workflows by using ECG data to flag patients who should receive further testing, potentially enabling earlier diagnosis and management.

The press release cites multiple references on PH detection and related studies, reinforcing the evidence base behind the technology.

Clinical development drew on training with over 250,000 de-identified Mayo Clinic records, with independent analyses showing 73% sensitivity and 74.4% specificity in a 21,066-patient multi-center study, plus real-world analyses indicating strong detection for PAH and CTEPH.

Independent real-world data suggest ECG-AI identified over 85% of patients with pulmonary arterial hypertension and about 78% with chronic thromboembolic PH, indicating potential for earlier detection of treatable subgroups.

The algorithm was developed using Mayo Clinic data and demonstrates 73% sensitivity and 74.4% specificity in a large dyspnea cohort, underscoring robustness across centers.

The AI solution integrates with electronic health record systems and ECG management platforms and operates entirely within the health system environment, with no patient data leaving the hospital.

As an SaMD, the system detects early signs of PH and integrates with EHRs and ECG platforms, maintaining in-system operation for real-time decision support.