Tempus AI has earned U.S. Food and Drug Administration (FDA) 510(k) clearance for Tempus ECG-AF, an algorithm that leverages artificial intelligence (AI) to identify patients at a higher risk of atrial fibrillation/flutter (AF).
This FDA nod represents the first for an AF indication in the category of “cardiovascular machine learning-based notification software.” The AI-based clinical tool aims to earlier identify cardiovascular diseases and conditions.
The algorithm analyzes recordings of 12-lead ECG devices and detects signs related to experience of AF in the next year. It’s meant for use on resting 12-lead ECG recordings gathered at a healthcare facility from patients aged 65 and up without a known history of AF or other specified conditions.
Tempus AI advised ECG-AF should be supported by the patient’s original ECG recordings and other tests, symptoms, and clinical history. The algorithm doesn’t describe overall risk of experiencing AF and shouldn’t be the sole basis for diagnosing it, the company said. ECG-AF results should also not be used as the basis for AF treatment or to rule out AF follow-up.
“We believe that there are too many patients who die from AF-related causes unnecessarily due to underdiagnosis or undertreatment,” said Brandon Fornwalt, MD, Ph.D., Tempus’ senior VP of cardiology. “The clearance of Tempus ECG-AF allows us to better support clinicians in potentially finding patients much earlier, when treatments can be more effective.”