Key Points
- The FDA has granted 510(k) clearance to Tempus AI for ECG-PH, an artificial intelligence software that identifies pulmonary hypertension indicators from 12-lead electrocardiograms.
- ECG-PH represents Tempus’ third cardiovascular AI solution to receive FDA approval, following clearances for atrial fibrillation and low ejection fraction detection tools.
- Approximately 1% of the general population and as many as 10% of individuals over age 65 are affected by pulmonary hypertension.
- The software serves as a screening aid rather than a definitive diagnostic instrument and requires integration with additional clinical data.
- ECG-PH is authorized exclusively for patients 40 years and older presenting cardiovascular symptoms without existing PH diagnoses.
The FDA has granted 510(k) clearance to Tempus AI (TEM) for its ECG-PH software, an artificial intelligence platform that identifies indicators of pulmonary hypertension through analysis of standard 12-lead electrocardiogram readings.
This regulatory approval represents the company’s third FDA-cleared cardiovascular AI solution. Tempus previously secured clearances for diagnostic tools focused on detecting atrial fibrillation and identifying low ejection fraction.
Pulmonary hypertension develops when lung blood vessels become narrowed and rigid, creating excessive strain on the heart’s right ventricle. The condition impacts approximately 1% of people worldwide and affects up to 10% of the elderly population above 65 years.
The condition often goes undetected due to its subtle presentation. Initial manifestations like tiredness and breathing difficulties are frequently dismissed, while the gold-standard diagnostic method requires right heart catheterization, an invasive cardiac procedure.
Tempus ECG-PH addresses this diagnostic gap. The platform evaluates resting 12-lead ECG data from non-ambulatory recordings and identifies markers associated with elevated mean pulmonary artery pressure, a key physiological indicator of PH.
Target Patient Population
The software is intended for patients 40 years of age or older presenting with cardiovascular manifestations including dyspnea, fatigue, chest discomfort, or edema, who have not been previously diagnosed with pulmonary hypertension.
ECG-PH delivers a binary determinationāeither detecting the indicator or not. Healthcare providers must integrate this output with comprehensive patient assessment data.
Tempus emphasizes that the software functions as an adjunctive screening tool rather than a replacement for clinical judgment. Physicians must evaluate results in conjunction with the source ECG recording, patient symptomatology, diagnostic history, and additional clinical information.
The platform is not designed for ongoing patient monitoring and cannot analyze electrocardiograms displaying paced cardiac rhythms.
Expanding Cardiac AI Capabilities
Dr. Brandon Fornwalt, Senior Vice President of Cardiology at Tempus, indicated that this clearance advances the company’s mission to enable earlier cardiovascular risk detection through artificial intelligence.
“Historically, pulmonary hypertension has evaded early diagnosis due to vague initial symptoms and the invasive nature of definitive testing,” Dr. Fornwalt explained.
He emphasized the organization’s objective to assist physicians in recognizing at-risk patients “throughout the continuum of cardiovascular disease.”
Tempus maintains one of the world’s most extensive multimodal healthcare data repositories and deploys AI solutions across diverse medical specialties beyond cardiology.
The ECG-PH authorization strengthens a cardiovascular AI suite that now addresses three separate conditions, each utilizing the widely available 12-lead ECG as the data source.
The 510(k) clearance pathway applies to medical devices demonstrating substantial equivalence to previously authorized products, representing a standard regulatory route for software-based diagnostic applications.
Tempus ECG-PH has received authorization for deployment in healthcare settings using clinical-grade diagnostic electrocardiogram recordings.





