AI-Enabled Digital Stethoscope Detects Pregnancy-Related Heart Disease

Peripartum cardiomyopathy, a heart failure condition affecting women in late pregnancy or post-pregnancy, significantly impairs the heart's ability to pump blood. While many recover normal heart function with treatment, it can escalate to severe heart failure in some cases, making early detection vital. However, diagnosis is often challenging as it typically occurs at the end of pregnancy or in the months after, with symptoms mirroring those of normal pregnancy, such as breathlessness and swelling in the feet and legs. Now, new research has revealed that screening for this condition using an AI-powered digital stethoscope in conjunction with an electrocardiogram (EKG) may be twice as effective in detection compared to standard obstetric care, which includes clinical EKGs.

In a study led by Mayo Clinic (Jacksonville, FL, USA) that involved nearly 1,200 Nigerian women, either pregnant or postpartum, researchers found that peripartum cardiomyopathy was identified twice as frequently when EKG tests were conducted using a digital stethoscope equipped with an AI algorithm, as opposed to using clinical EKGs alongside routine obstetric care. Nigeria has the highest reported incidence of peripartum cardiomyopathy worldwide, impacting as many as 1 in 96 pregnancies. The AI screening utilized a digital stethoscope to record the heart's electrical activity and sounds. Women in the AI intervention group also received an initial echocardiogram to confirm the AI algorithm's effectiveness.

The AI-enhanced stethoscope employed an algorithm initially designed for 12-lead EKG data, adapted for use with a single-lead EKG via a digital stethoscope. This adaptation allowed for the prediction of the likelihood of left ventricular dysfunction. The control group underwent traditional EKG tests without the AI digital stethoscope, and health professionals reviewed their results. Echocardiograms were used to measure left ventricular ejection fraction, a key indicator of the heart's pumping efficiency, typically ranging between 55% and 70%. In this study, a left ventricular ejection fraction below 50% was indicative of cardiomyopathy. The AI-enabled stethoscope group had a 4% detection rate of pregnancy-related cardiomyopathy, compared to just 1.8% in the control group. This disparity suggests that standard care might miss about half of all cardiomyopathy cases.

“We demonstrated for the first time in an obstetric population that AI-guided screening using a digital stethoscope improved the diagnosis of this potentially life-threatening and treatable condition,” said lead study author Demilade A. Adedinsewo, M.D., M.P.H. “This research can change current clinical practice from one that is reactive and symptom-driven to a more proactive approach of identifying pregnancy-related cardiac dysfunction using a simple, low-cost and effective screening tool. Earlier diagnosis would facilitate prompt and appropriate management of cardiomyopathy and reduce associated disease and death.”

Related Links:
Mayo Clinic