ROCHESTER, Minn. — Every year, some mothers die after giving birth due to heart problems, and many of these deaths could be prevented. The ability to screen for heart weakness before pregnancy could play a crucial role in identifying women who may need additional care to improve pregnancy outcomes. Mayo Clinic researchers, led by Anja Kinaszczuk, D.O., and Demilade Adedinsewo, M.D., tested artificial intelligence (AI) tools, using recordings from an electrocardiogram (ECG) and a digital stethoscope, to find unknown heart problems in women of childbearing age seen in primary care.
Study findings[1] published in the Annals of Family Medicine show high diagnostic performance of these technologies to detect left ventricular ejection fraction below 50%, indicating heart muscle weakness. These tools were tested on two groups of women aged 18 to 49.
Group 1: 100 women already scheduled for an echocardiogram (the best test to evaluate heart muscle function). They also had a standard clinical ECG and digital stethoscope recording of the heart’s electrical activity and heart sounds.
Group 2: 100 women seen for routine primary care visits to see how often the AI tools would find heart problems.
The AI-ECG demonstrated an area under the curve (AUC) of .94 while the AI digital stethoscope, Eko DUO, achieved an even higher AUC of 0.98, indicating strong diagnostic accuracy. In the second cohort, the prevalence of positive AI screening results was 1% for the AI-ECG and 3.2% for the AI-stethoscope.
"Statistically, nearly half of pregnancies in this country are unplanned, and approximately 1% to 2% of women may have heart problems they don't know about", says Dr. Adedinsewo, a cardiologist and senior author of the study.
"Our research findings suggest that these AI tools could be used to screen women before pregnancy, allowing for improved pregnancy planning and risk stratification, early treatment, and better health outcomes which addresses a critical gap in current maternal care."Dr. Adedinsewo, cardiologist, senior author of study
This research builds upon earlier published studies, including a pilot prospective study[2] evaluating AI digital tools to detect pregnancy-related cardiomyopathy among obstetric patients in the U.S. and a pragmatic randomized clinical trial[3] of women in Nigeria who were pregnant or had recently given birth. Collectively, this research highlights the potential of AI to modernize cardiovascular screening, enabling earlier identification and management of heart muscle weakness in women of reproductive age. Further research is underway to explore the potential of using these technologies to screen for heart weakness in broader populations.
Mayo Clinic has licensed the underlying technology to EKO Health for its digital stethoscope with embedded ECG electrodes and to Anumana for the 12-lead ECG. Mayo Clinic and some study authors have a financial interest in this technology. Mayo Clinic will use any revenue it receives to support its not-for-profit mission in patient care, education and research.
References
1) https://doi.org/10.1370/afm.230627
2) https://pubmed.ncbi.nlm.nih.gov/38989045/
3) https://www.nature.com/articles/s41591-024-03243-9
(Newswise/VKR)