Anumana's Breakthrough ECG-AI Algorithm Achieves FDA Approval for Low Ejection Fraction Detection
Anumana, Inc., a prominent player in the field of AI-driven health technology, has announced a significant achievement in partnership with Mayo Clinic. They have received FDA 510(k) clearance for their revolutionary medical device, ECG-AI LEF, which leverages artificial intelligence (AI) to identify low ejection fraction (LEF) in patients at risk of heart failure.
LEF, which often exhibits no noticeable symptoms, is a crucial indicator of heart failure. With the increasing prevalence of heart failure and the associated challenges, including morbidity, mortality, rehospitalizations, and economic burdens, there is a pressing need to detect and manage LEF in patients.ECG-AI LEF is a pioneering software-as-a-medical device (SaMD) designed for the screening of LEF in adults at risk of heart failure. It analyzes data from standard 12-lead electrocardiograms (ECGs), a widely used diagnostic tool in both primary and specialized healthcare settings.
The clinical validation of ECG-AI LEF was remarkable. In a retrospective clinical study involving 16,000 patients from diverse racial backgrounds, the device exceeded expectations. It achieved an impressive 84.5% sensitivity and 83.6% specificity, meeting its primary endpoint. Additionally, ECG-AI LEF demonstrated an AUROC of 0.932, indicating its exceptional ability to distinguish between LEF and an ejection fraction (EF) exceeding 40%. This performance outperforms many current tests in standard heart failure care.
Anumana is committed to driving the adoption of the ECG-AI category and is actively engaged in the clinical development and commercialization of innovative healthcare technologies. The recently cleared ECG-AI LEF can be seamlessly integrated into various ECG information management systems or directly into a patient's electronic health record using Anumana's web-based ECG Viewer. This integration is set to enhance clinical decision-making and advance the field of cardiology.