AI-Enhanced ECG Screens for Heart Failure Risk in Resource-Limited Settings

In sub-Saharan Africa, patients often develop the disease at younger ages and face worse outcomes. Early identification of left ventricular systolic dysfunction, a key precursor to heart failure, is limited by scarce and costly echocardiography. A new study shows an artificial intelligence approach can enable earlier, lower-cost screening in routine care settings.

Researchers at UT Southwestern Medical Center (Dallas, TX, USA) evaluated an artificial intelligence–augmented electrocardiogram (AI-ECG) that analyzes standard electrocardiograms to screen for left ventricular systolic dysfunction (LVSD) and other antecedents of heart failure. The investigation focused on Kenya, where access to echocardiography is constrained. The AI-ECG was deployed across eight health care facilities to test its performance as a front-line screening tool. The goal was to determine whether ECG data, enhanced by artificial intelligence, could reliably flag patients for confirmatory imaging.

Nearly 6,000 patients seeking routine clinical care received AI-ECG screening. A subset of 1,444 patients also underwent echocardiography to verify AI-ECG findings. Among those who received echocardiograms, the algorithm identified LVSD in 14.1% of patients. The negative predictive value was 99.1%, indicating that nearly all patients with a negative screen were confirmed negative on echocardiography.

The algorithm showed high sensitivity, correctly identifying 95.6% of patients with LVSD, and demonstrated 79.4% specificity for excluding those without the condition. Positive AI-ECG results were strongly associated with other markers of adverse cardiac remodeling, including left ventricular hypertrophy and diastolic dysfunction. The authors concluded that AI-ECG supports systematic screening for LVSD in settings where widespread echocardiography is not feasible, offering a potential low-cost pathway to earlier detection.

Findings were published in JAMA Cardiology on May 6, 2026. Collaborators included clinical partners in Kenya, notably M.P. Shah Hospital, with patient recruitment spanning eight facilities. The study underscores the potential for artificial intelligence to extend the reach of cardiac screening in resource-limited environments while reserving echocardiography for patients flagged at higher risk.

“These findings support AI-ECG as a practical, scalable screening tool that can effectively identify individuals at risk for heart failure in resource-limited settings where access to echocardiography is constrained, addressing a critical gap in global cardiovascular care,” said Ambarish Pandey, M.D., Associate Professor of Internal Medicine in the Division of Cardiology and in the Peter O’Donnell Jr. School of Public Health at UT Southwestern.

Related Links
UT Southwestern Medical Center