Artificial Intelligence and Cardiac Equity in India: Can Technology Bridge the Care Divide?
Year: 2025
Type: article
Abstract: Cardiovascular disease (CVD) continues to be the leading cause of death globally, accounting for nearly 18 million fatalities each year.This alarming trend is particularly pronounced in low-and middleincome countries (LMICs), including India, where the incidence of CVD is on the rise. 1 India confronts a distinctive "double burden" characterized by a high incidence of both communicable and noncommunicable diseases, compounded by significant disparities in healthcare access.While metropolitan regions are equipped with state-of-the-art cardiac institutes; rural and tribal communities often find themselves inadequately served, facing limitations in diagnostic facilities, a scarcity of specialists, and low levels of health literacy.2 In this context, the implementation of artificial intelligence (AI)enhanced cardiac screening presents a potentially transformative intervention.Such technological advancements have the capacity to facilitate early, scalable, and cost-effective diagnoses, thereby democratizing access to high-quality cardiovascular care for India's most underserved populations.3,4 The Promise o f Ai i n indiAn CArdiAC sCreeningArtificial intelligence tools are increasingly demonstrating diagnostic accuracy comparable to human experts in various cardiovascular tasks, including electrocardiogram (ECG) interpretation, echocardiography, and risk prediction modeling.For example, an AI-based cardiovascular disease (AICVD) risk score developed in India showed higher predictive performance for cardiovascular events over 10 years than conventional scores like Framingham and QRisk3 in cohorts from Apollo Hospitals, with an area under the curve (AUC) around 0.85. 5 In India, there are early efforts to adapt such technologies to local needs, although many remain in pilot or development phases:• AI-enabled ECG interpretation: Narayana Health, partnering with Medha AI, has developed a model that predicts left ventricular ejection fraction from a standard ECG image in near-real time, enabling rapid identification of possible heart failure, especially in resource-limited settings.6 • Portable echocardiography for rheumatic heart disease (RHD):A recent study of 511 echocardiograms in children (282 with RHD, 229 normal) focused on mitral regurgitation, using deep learning and image localization techniques to detect RHD with high accuracy (AUC ≈ 0.93).7 Also, the "RHD Roko" pilot in Uttar Pradesh includes screening schoolchildren using AI-enabled digital stethoscopes plus echocardiography at district hospitals.8 • Data-driven risk stratification: The AICVD model (noted above) used electronic health record data from ~31,600 participants in India, integrating many risk factors, and was validated both internally and externally, outperforming Framingham and QRisk3 in Indian populations.5 These initiatives suggest that AI has strong potential to decentralize cardiac diagnostics beyond Tertiary Centers, improve early detection, and reduce preventable mortality.However, much of this is still emerging work; wider validation, regulatory approval, cost-effectiveness, and integration into health systems remain to be fully addressed.
Author: Richa G Thaman
Language: English
Subfield: Health Informatics
Field: Medicine
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W4416774755