Anumana ECG-AI LEF
Anumana
Analyzes a standard 12-lead ECG to identify patients likely to have reduced left ventricular ejection fraction, as a prompt for confirmatory echocardiography.
At A Glance
- Manufacturer
- Anumana
- Medical specialty
- Cardiology
- Clearance route
- 510(k)
- Year authorized
- 2022
- Modality
- ECG
- Role in workflow
- Screening
This is one of the more genuinely surprising results in clinical machine learning. The ECG was not thought to contain reliable information about ejection fraction, and cardiologists cannot extract it by eye. A model trained on paired ECG and echo data can, which means the signal was always there and human pattern recognition simply could not reach it.
Where It Is Used
- Opportunistic heart failure screening
- Identifying candidates for confirmatory echocardiography
- Population screening from existing ECG data
Limitations Worth Knowing
The output is a screening flag, not a measurement, and it requires echocardiographic confirmation. Positive predictive value depends entirely on the prevalence of reduced ejection fraction in the screened population, so applying it broadly produces mostly false positives.
Regulatory status, indications, and labeling change over time. Confirm current status on the FDA's AI/ML-Enabled Medical Devices list and in the manufacturer's current instructions for use before relying on anything stated here.