In a first-of-its-kind breakthrough for India’s healthcare sector, Narayana Health has developed an artificial intelligence (AI) model that can detect heart failure from a standard ECG image in just 10 seconds. Created by the hospital’s Clinical Research Team in collaboration with its AI division, Medha AI, the tool predicts left ventricular ejection fraction (EF) from ECG scans, enabling early detection of heart failure in near real-time.
Addressing Diagnostic Gaps Nationwide
The AI model aims to close critical diagnostic gaps for over 10 million Indians living with heart failure, particularly in regions lacking access to echocardiography—the current standard for diagnosis. Trained on more than 100,000 ECG-image and echo-report pairings, it demonstrated 97% accuracy in identifying patients with severely reduced EF (≤35%) across a multi-centre validation cohort of 57,000 patients.
A Game-Changer for Accessible Cardiac Care
Dr. Devi Shetty, Founder and Chairman of Narayana Health, called the innovation a milestone in accessible diagnostics. “This is not just a model, it’s a demonstration of India’s ability to combine clinical excellence with digital innovation. Being able to predict heart function from a simple ECG is a game-changer in making advanced diagnostics truly accessible to all,” he said.
Seamless Integration into Clinical Systems
The AI has been integrated into Athma, Narayana Health’s EMR system, and is undergoing validation at the Narayana Institute of Cardiac Sciences, Mazumdar Shaw Medical Center, and other centres in Bengaluru and Kolkata. Early trials suggest the tool can flag potential heart failure up to 58 days before an echocardiogram is typically ordered, giving doctors valuable time for intervention.
Breaking Barriers for Rural Deployment
Highlighting its unique adaptability, Dr. Pradeep Narayan, Chief Scientific Officer, noted, “The tool works even from a mobile photo of an ECG strip. That opens possibilities for deployment in rural and remote areas where resources are scarce but the need is high.” This capability positions the AI as a viable solution for mobile medical units and community health programs.
Global Recognition and Clinical Validation
As reported by bwhealthcareworld, the innovation has earned accolades at international conferences such as HRX (Atlanta) and APHRS (Sydney). It has also been published in the Indian Heart Journal. It also secured the BMJ South Asia Award for Digital Innovation. Dr. Deepak Padmanabhan, Senior Consultant, stressed, “Heart failure often goes undiagnosed due to lack of tools or expertise. This AI provides reliable support even in low-resource environments.”
Future Roadmap for AI-Driven Cardiac Care
Looking ahead, Narayana Health plans to expand the tool’s capabilities to detect other cardiac abnormalities and early signs of coronary artery disease. The roadmap includes obtaining regulatory approvals and making the tool open access for broader adoption.
Embedding AI Across the Healthcare Ecosystem
Led by Vivek Rajagopal, Group Chief Analytics & AI Officer, Narayana Health’s AI initiative is integrating such tools across its ecosystem. These range from diagnostic support and automated documentation to predictive supply chain and resource planning. Emphasising the importance of equitable innovation, Dr. P. M. Uthappa, Group Chief Medical Director, stated, “Responsible AI in healthcare isn’t just about innovation. It’s about fairness, transparency, and ensuring the most underserved also benefit from what’s possible.”