Narayana Health’s Clinical Research team, in collaboration with Medha AI, its advanced analytics and AI division, has developed a pioneering artificial intelligence model capable of predicting left ventricular ejection fraction (LVEF)—a key indicator of heart function—using a standard ECG image in near real-time.
This cutting-edge model is designed for seamless integration into clinical workflows, enabling early detection of heart failure and expanding diagnostic access, particularly in resource-limited settings where echocardiography (ECHO) is not readily available.
With heart failure impacting nearly 10 million Indians and contributing to over 1.8 million hospitalisations annually, the AI tool bridges a crucial diagnostic gap. It allows scalable, early screening by leveraging widely available ECG machines and cloud-based analytics, eliminating sole reliance on ECHO.
Trained on over 100,000 ECG images paired with ECHO reports, the model demonstrated high accuracy during external validation involving 57,000+ patients across 14 tertiary centres. It successfully identified 97% of cases with severely reduced EF (≤35%), often flagging these patients an average of 58 days before ECHO was eventually ordered—offering significant lead time for intervention.
At the unveiling event, key leadership including Dr. Emmanuel Rupert (Managing Director & Group CEO), Dr. Pradeep Narayan (Chief Scientific Officer and Lead Researcher), Dr. PM Uthappa (Group Chief Medical Director), and Dr. Deepak Padmanabhan (Senior Consultant, Strategic Lead – Cardiac EP Collegium) were in attendance.
Dr. Devi Shetty, Founder and Chairman of Narayana Health, highlighted the significance:
“This is more than a tech breakthrough—it’s a milestone in making heart care universally accessible. Using AI to predict heart failure from a simple ECG empowers clinicians with faster decisions and broader reach. This is India’s moment to lead global healthcare innovation.”
The AI model is now integrated into Athma, Narayana Health’s homegrown Electronic Medical Record (EMR) platform. It is undergoing real-world validation at Narayana Institute of Cardiac Sciences, Mazumdar Shaw Medical Centre, and Rabindranath Tagore International Institute of Cardiac Sciences, along with select clinics in Bengaluru.
Designed to operate in near real-time, the tool aims to support early triage and clinical decision-making without disrupting existing provider workflows.