Anumana Receives U.S. FDA 510(k) Clearance for ECG-AI Algorithm to Detect Low Ejection Fraction

CAMBRIDGE, Mass.–(BUSINESS WIRE)–Anumana, Inc., a leading AI-driven health technology and nference portfolio company working in collaboration with Mayo Clinic, today announced U.S. Food and Drug Administration (FDA) 510(k) clearance for ECG-AI LEF, a breakthrough artificial intelligence (AI)-powered medical device to detect low ejection fraction (LEF) in patients at risk of heart failure.

“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”

Tweet this

LEF, or a weak heart pump, is a significant, at times asymptomatic, and commonly undiagnosed indicator of heart failure.¹ The increasing prevalence of heart failure and its associated morbidity, mortality, rehospitalizations, and societal costs^2,3 underscores the need to identify and manage patients with LEF.

“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,” said Paul Friedman, M.D., chair of the Department of Cardiovascular Medicine at Mayo Clinic in Rochester, Minn. and chair of Anumana’s Board of Advisors. “It allows identification of otherwise hidden disease, for which many effective, lifesaving treatments are available – once the presence of the disease is known.”

Developed in partnership with Mayo Clinic, Anumana’s ECG-AI LEF is an innovative software-as-a-medical device (SaMD) designed to screen for LEF in adults at risk for heart failure using data from a routine 12-lead electrocardiogram (ECG), a rapid and common test used in both primary and specialty care. Based on pioneering research from Mayo Clinic,⁴ the algorithm has been developed utilizing over 100,000 ECG and echocardiogram data pairs from unique patients and has been clinically tested in more than 25 studies involving more than 40,000 patients in the U.S. and internationally.⁵

Anumana’s ECG-AI LEF was clinically validated in a multi-site, retrospective clinical study of 16,000 racially diverse patients, achieving its primary endpoint with an 84.5% sensitivity and 83.6% specificity.⁶ ECG-AI LEF achieved an AUROC⁷ of 0.932, demonstrating an ability to differentiate between LEF and ejection fraction (EF) >40% extremely well (a score ≥0.90 is considered excellent⁸) and better than most tests currently used in heart failure standard of care.⁹

Additionally, the EAGLE study, a groundbreaking prospective, randomized, controlled clinical trial by Mayo Clinic evaluated the use of an investigational version of the algorithm in routine clinical care of 22,641 adults by 120 primary care teams from 45 clinics or hospitals, demonstrating that ECG-AI LEF implementation improved clinician’s ability to diagnosis of LEF by 31% versus standard of care without increasing the overall rate of echocardiogram usage.¹⁰

ECG-AI LEF is one of Anumana’s broad pipeline of algorithms, including three additional FDA breakthrough device designation algorithms (pulmonary hypertension, cardiac amyloidosis, and hyperkalemia), and is founded on more than six years of pioneering ECG-AI research and development at Mayo Clinic, including nearly 100 peer-reviewed publications to date.

“Anumana was established in 2021 by nference in partnership with Mayo Clinic to unlock the electrical language of the heart through deep learning and improve disease diagnosis and patient care,” said Murali Aravamudan, co-founder and CEO of Anumana and nference. “In the short time of two years we have secured multiple FDA breakthrough device designations, entered multi-year agreements with three pharma partners, successfully established two new medical procedure codes for ECG AI technology, and now achieved our first FDA breakthrough medical device clearance. This is a significant milestone, and we are excited about the next phase of the journey, deploying our technology in the U.S. and globally to empower clinicians and enhance real-world clinical care.”

Anumana is focused on driving fast-paced adoption of the new ECG-AI category, clinically developing and commercializing its novel technologies in healthcare. The newly FDA-cleared ECG-AI LEF can be easily integrated with various ECG information management systems or directly with a patient’s electronic health record via Anumana’s web-based ECG Viewer to support clinical decision-making.

Anumana spearheaded the effort to bring reimbursement to ECG-AI, receiving approval for two Category III CPT® codes from the American Medical Association in 2022. These codes are now available and designed to facilitate the use, adoption, and potential reimbursement of emerging technologies in clinical workflows.

To learn more about ECG-AI LEF and schedule a demo, visit us at anumana.ai.

About Anumana

Anumana is a leading AI-driven health technology company leveraging cutting-edge AI and industry-leading translational science to unlock the electrical language of the heart as never before. The company was founded by nference in partnership with Mayo Clinic to leverage the clinical and technical expertise of both organizations to develop innovative ECG-AI technology from an investigational resource to a clinically meaningful tool for clinicians to advance patient care. Anumana’s software-as-a-medical device (SaMD) solutions aim to detect hidden diseases using ECG-AI algorithms to enhance and improve interventional procedures through real-time AI insights.

Anumana was named one of Fierce MedTech’s Fierce 15 companies for 2022. Follow Anumana at anumana.ai and on LinkedIn and Twitter.

Dr. Friedman and Mayo Clinic have a financial interest in the technology referenced in this press release. Mayo Clinic will use any revenue it receives to support its not-for-profit mission in patient care, education, and research.

References

1. Jaskanwal D Sara, Takumi Toya, Riad Taher, Amir Lerman, Bernard J Gersh, Nandan S Anavekar. Asymptomatic Left Ventricle Systolic Dysfunction. European Cardiology Review 2020, 15:e13; https://doi.org/10.15420/ecr.2019.14.
2. Tsao, C.W., Aday, A.W., Almarzooq Z.I., et al. Heart Disease and Stroke Statistics—2023 Update: A Report From the American Heart Association. Circulation Vol. 147, No. 8; https://www.ahajournals.org/doi/10.1161/CIR.0000000000001123#d330256e1.
3. Khazanie P, Allen LA. Systematizing Heart Failure Population Health. Heart Fail Clin. 2020 Oct;16(4):457-466. Doi: 10.1016/j.hfc.2020.06.006. Epub 2020 Jul 21. PMID: 32888640; PMCID: PMC7737815; https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7737815/.
4. Attia, Z.I., Kapa, S., Lopez-Jimenez, F. et al. Screening for cardiac contractile dysfunction using an artificial intelligence–enabled electrocardiogram. Nat Med 25, 70–74 (2019); https://doi.org/10.1038/s41591-018-0240-2.
5. Anumana data on file and published studies.
6. Anumana data on file (NCT04963218).
7. Area Under the Receiving Operating Characteristic curve (AUROC) is a performance metric commonly used to evaluate the ability of a model to distinguish between two classes. An AUROC of 0.5 suggests that the model is no better than random and an AUROC of 1.0 indicates a perfect classifier; https://en.wikipedia.org/wiki/Receiver_operating_characteristic.
8. Muller MP, Tomlinson G, Marrie TJ, Tang P, McGeer A, Low DE, et al. Can routine laboratory tests discriminate between severe acute respiratory syndrome and other causes of community-acquired pneumonia? Clin Infect Dis. 40:1079–86 (2005); https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8831439/.
9. Such as NT-proBNP, a common blood test used to help identify LEF today with an AUROC of 0.83; Koschack, J., Scherer, M., Lüers, C. et al. Natriuretic peptide vs. clinical information for diagnosis of left ventricular systolic dysfunction in primary care. BMC Fam Pract 9, 14 (2008); https://doi.org/10.1186/1471-2296-9-14h.
10. Yao, X., Rushlow, D.R., Inselman, J.W. et al. Artificial intelligence–enabled electrocardiograms for identification of patients with low ejection fraction: a pragmatic, randomized clinical trial. Nat Med 27, 815–819 (2021); https://doi.org/10.1038/s41591-021-01335-4.