Each year, heart-related complications contribute to maternal mortality, particularly after childbirth. Early detection of heart weaknesses in women of reproductive age could play a vital role in preventing these tragic outcomes. Recent research led by Mayo Clinic researchers, including Dr. Anja Kinaszczuk and Dr. Demilade Adedinsewo, is exploring the potential of artificial intelligence (AI) tools to improve screening processes for heart problems before pregnancy.
Using recordings from electrocardiograms (ECG) and digital stethoscopes, the AI tools have demonstrated high accuracy in identifying women with a left ventricular ejection fraction below 50%, a key indicator of heart muscle weakness. The study involved two groups of women aged 18 to 49: one with prior echocardiogram testing and another from routine primary care visits.
The AI-ECG system achieved an area under the curve (AUC) of 0.94, while the AI-powered digital stethoscope, Eko DUO, reached an impressive AUC of 0.98, indicating strong diagnostic performance. These findings suggest that such technology could be a game-changer for early screening, especially in settings where access to traditional testing is limited.
“Approximately 1% to 2% of women may have undiagnosed heart problems, and almost half of pregnancies are unplanned,” explains Dr. Adedinsewo. “Our research indicates these AI tools could be invaluable for screening women before conception, enabling better pregnancy planning, early intervention, and improved health outcomes.”
This research builds on previous studies, including a pilot project evaluating AI tools for pregnancy-related cardiomyopathy detection and a pragmatic trial in Nigeria involving women who recently gave birth. The collective findings underscore the potential of AI in modernizing cardiovascular screening in women of reproductive age.
Mayo Clinic has licensed this innovative technology to companies such as EKO Health and Anumana, and the revenue will support ongoing patient care, education, and research efforts. The development and deployment of these tools mark a significant step forward in maternal health and cardiovascular medicine.
For more details, the full study can be accessed in the Annals of Family Medicine.
