AI-Guided Echocardiography Enables Mobile Cardiac Imaging in Rural Communities

Delayed or unavailable imaging can hinder detection of serious heart abnormalities that require follow-up. Researchers have now shown that AI-guided echocardiography combined with off-site expert review may help expand access to mobile heart imaging in rural settings.

The work was conducted through the RURAL (Risk Underlying Rural Areas Longitudinal) study and involved the Duke Clinical Research Institute Imaging Core Laboratory. The approach used AI-driven software to guide nonsonographer healthcare professionals as they acquired echocardiographic images in a mobile health unit. Experienced sonographers and echocardiography-certified physicians then reviewed the images remotely for completeness, image quality, and cardiac measurements.

The study included 4,597 adult volunteers aged 25 to 65 who underwent scanning in 10 rural counties across Alabama, Kentucky, Louisiana, and Mississippi between November 2021 and June 2026. Fourteen nonsonographers, including nurses, nursing assistants, and computed tomography technologists, completed a two-hour training program with virtual and hands-on components. Each participant’s broader study visit lasted 3.5 hours, while the echocardiography workflow was evaluated against a target scan time of less than 15 minutes.

All six required standard echocardiography views were obtained in 98% of participants. Core laboratory experts rated the studies adequate for cardiac measurements more than 99% of the time, while diagnostic-quality rates for individual views ranged from 77% to 94%. Scans were completed within the 15-minute target window in 81% of cases, with a median duration of 9.5 minutes.

Critical findings were identified in 133 scans. The most common was an ejection fraction below 50%, indicating reduced pumping function of the heart’s main chamber. Among participants with critical findings, 99% received their results within three days.

The preliminary study is scheduled for presentation at the American Heart Association’s Scientific Sessions 2026 in Chicago from November 6 to 9. Limitations included the absence of a direct comparison with sonographer-acquired images, internal-only reproducibility assessment, no systematic evaluation of operator variability, use of a single software platform, and reliance on one core laboratory. The study also did not assess whether returning results led to changes in patient care or health outcomes.

“We were surprised at how quickly and well the nurses and CT technologists who had never done a heart ultrasound were able to adapt to this technology and produce quality images for review by experts,” said study author Dr. Sanjana Ravi, an internal medicine and pediatrics resident at the Duke University School of Medicine.

“With real-time AI guidance and a rigorous quality-control system behind it, a nurse or CT technologist in a mobile unit can produce heart images good enough for research and good enough to flag serious heart problems that need follow-up, bringing the imaging to the community rather than asking the community to travel to it,” added Dr. Ravi.

Related Links
Duke University School of Medicine