Imagine a future where heart disease could be predicted years before symptoms appear, allowing for early intervention and potentially saving countless lives. This is no longer science fiction—it's the reality being shaped by groundbreaking advancements in AI-driven cardiovascular science. Anumana, a trailblazer in AI-powered diagnostics, is at the forefront of this revolution, and their latest revelations at the American Heart Association (AHA) Scientific Sessions 2025 are nothing short of transformative.
But here's where it gets controversial: Can AI truly outperform traditional clinical methods in predicting heart failure? Anumana's late-breaking study, published simultaneously in the Journal of the American College of Cardiology, suggests it can. By applying AI to electrocardiograms (ECG-AI), researchers found that this technology significantly enhances the near-term prediction of heart failure, outperforming established clinical risk models. This isn’t just an incremental improvement—it’s a paradigm shift.
The study, titled “Enhanced Prediction of Incident Heart Failure Using Artificial Intelligence-Driven Analysis of 12-Lead Electrocardiogram Waveforms: A HeartShare/AMP-HF Pooled Cohort Analysis,” analyzed data from over 14,000 participants across three major longitudinal cohorts: the Framingham Heart Study, Multi-Ethnic Study of Atherosclerosis, and Cardiovascular Health Study. The results were striking: integrating Anumana’s ECG-AI with the PREVENT-HF clinical risk equation reclassified up to 12.5% of individuals into higher-risk categories that traditional methods missed. Even more astonishing, participants with positive ECG-AI results were over 20 times more likely to develop heart failure within three years compared to those with negative results.
And this is the part most people miss: ECG-AI detects subtle electrical changes in the heart that signal early cardiac dysfunction—changes that often go unnoticed by conventional methods. As Akshay S. Desai, MD, MPH, lead investigator and Director of the Heart Failure Disease Management Program at Brigham and Women’s Hospital, explains, “The implication is that ECG-AI may help clinicians identify at-risk patients years before symptoms of heart failure appear, creating opportunities to start preventive therapy sooner and improve long-term outcomes.”
This study was made possible through collaboration with the National Heart, Lung, and Blood Institute’s HeartShare/AMP Heart Failure Program and the BioData Catalyst platform, which provided access to deeply phenotyped, longitudinal cohorts and robust analytic capabilities. This rigorous evaluation underscores the potential of ECG-AI to revolutionize early heart failure risk prediction.
Simos Kedikoglou, MD, President and COO of Anumana, emphasizes the broader implications: “This publication marks meaningful progress toward a future where AI helps prevent disease rather than just detecting it. It demonstrates how our ECG-AI LEF algorithm can uncover early signs of heart failure, supporting clinicians in identifying at-risk patients sooner and enabling more proactive care.”
Anumana’s ECG-AI™ LEF algorithm achieved an impressive AUC of 0.944, with a sensitivity of 90.2% and specificity of 85.1%, showcasing its ability to detect patients at risk for heart failure. But Anumana didn’t stop there. At AHA 2025, they also presented three additional abstracts highlighting the versatility of AI across cardiovascular conditions:
- Multicenter Study of ECG-AI for Pulmonary Hypertension: Across five U.S. health systems, ECG-AI detected pulmonary hypertension with 84% sensitivity and 72% specificity, paving the way for earlier disease identification.
- AI ECG for Early Identification of Pulmonary Arterial and Chronic Thromboembolic Pulmonary Hypertension: A real-world data analysis revealed that over 74% of patients had at least one ECG flagged as positive by ECG-AI PH between initial symptom presentation and diagnosis, suggesting a potential reduction in diagnostic delays.
- Parity and Takotsubo Cardiomyopathy: Researchers developed a novel methodology to evaluate the association between the number of pregnancies and the risk of Takotsubo cardiomyopathy using electronic health record data.
These studies collectively reinforce Anumana’s leadership in translating advanced AI models into clinically meaningful solutions. Co-founded by nference and Mayo Clinic, Anumana is committed to transforming cardiovascular care through software-as-a-medical-device (SaMD) solutions that support early detection, clinical decision-making, and intraoperative guidance.
Here’s the thought-provoking question: As AI continues to advance, will it become the gold standard for cardiovascular diagnostics, or will it complement traditional methods? Share your thoughts in the comments below. To learn more about Anumana’s innovations, visit www.anumana.ai or explore their FDA-cleared ECG-AI™ LEF algorithm, now available in the U.S. and eligible for reimbursement as of January 2025. The future of heart health is here—and it’s powered by AI.
