AI-powered device to identify low LV ejection fraction gains FDA 510(k) clearance
Key takeaways:
- An AI-powered device to identify low left ventricular ejection fraction in patients at risk for HF gained 510(k) clearance from the FDA.
- The algorithm is designed to recognize low EF from a routine 12-lead ECG.
Anumana announced it received FDA 510(k) clearance for its novel artificial intelligence-powered device to detect low left ventricular ejection fraction in patients at risk for HF.
Developed in partnership with Mayo Clinic, the AI-powered device (ECG-AI LEF) uses data from a routine 12-lead ECG, according to a company press release.
The AI algorithm was trained using more than 100,000 ECG and echocardiogram data pairs from unique patients and was clinically tested in more than 25 studies including more than 40,000 patients in the U.S. and globally, according to the release.
Clinical validation of the ability of the device to identify low EF was conducted in a multi-site, retrospective clinical study of 16,000 patients, where the device achieved a specificity of 83.6%, a sensitivity of 84.5% and an area under the receiver operating characteristic curve of 0.932, according to the release.
Additionally, in the prospective, randomized, controlled EAGLE study conducted by Mayo Clinic, researchers demonstrated that the AI algorithm improved diagnosis of low EF by 31% compared with standard of care without increasing the overall rate of echocardiogram usage, according to the release.
“Anumana’s ECG-AI LEF fills an important unmet need — the lack of an easily accessible point-of-care, noninvasive and inexpensive tool to screen for a weak heart pump,” Paul Friedman, MD, chair of the department of cardiovascular medicine at Mayo Clinic and chair of Anumana’s board of advisors, said in the release. “It allows identification of otherwise hidden disease, for which many effective, lifesaving treatments are available — once the presence of the disease is known.”
Collapse