0
0
In a groundbreaking study presented at the Society for Maternal-Fetal Medicine’s (SMFM) annual meeting, The Pregnancy Meeting, researchers unveiled findings that suggest artificial intelligence (AI) can significantly aid clinicians in detecting congenital heart defects (CHDs) during routine prenatal ultrasounds.
Congenital heart defects—abnormalities of the heart present at birth—are the most common type of birth defect. According to the Centers for Disease Control and Prevention (CDC), approximately one in four babies born with a heart defect requires surgery or medical intervention within their first year of life. Despite advances in prenatal care, detecting CHDs during routine ultrasounds remains a challenge.
AI-Driven Improvement in Detection Rates
The study involved a cohort of physicians with experience ranging from one to more than 30 years, who reviewed 200 ultrasounds. A total of 14 OB-GYNs and maternal-fetal medicine (MFM) subspecialists assessed each ultrasound both with and without an AI-based software program. Researchers then compared the data to determine whether the software improved clinicians’ ability to detect CHDs.
The results were striking: irrespective of experience or subspecialty training, AI significantly enhanced clinicians’ ability to identify cases suspicious for congenital heart defects. The study also found that AI-assisted evaluations improved clinicians’ confidence levels and reduced the time required to assess potential CHDs.
Expert Insights on AI’s Potential
Dr. Jennifer Lam-Rachlin, the study’s lead author and a maternal-fetal medicine subspecialist, emphasized the importance of these findings.
“At least half of prenatal ultrasounds in the United States are examined by non-specialists—medical professionals, including OB-GYNs, who may not have specific training in prenatal ultrasound. This contributes to the suboptimal detection rates of congenital heart defects, even in developed countries like the U.S.,” said Lam-Rachlin, who also serves as the director of fetal echocardiography at Carnegie Imaging for Women and an assistant clinical professor at the Icahn School of Medicine at Mount Sinai West in New York City.
“Our findings show that AI-based software significantly improved detection rates not only among OB-GYNs but also among maternal-fetal medicine subspecialists. This has tremendous implications for neonatal outcomes and has the potential to change clinical practice.”
Christophe Gardella, Ph.D., chief technical officer for BrightHeart—the Paris-based manufacturer of the AI-based software—highlighted AI’s role in bridging the gap between specialists and non-specialists.
“Most congenital heart defects occur in pregnancies classified as low risk, meaning the pregnant individual is most likely seeing an OB-GYN rather than an MFM specialist. We designed our AI software in collaboration with experts to enhance detection rates among all practitioners, ensuring earlier diagnosis and better outcomes,” said Gardella.
Potential for Clinical Impact
With early detection playing a critical role in the management and treatment of congenital heart defects, AI-assisted evaluations could lead to timely interventions and improved neonatal health. If widely implemented, AI technology could help address existing disparities in prenatal care and ensure that more babies receive the necessary medical attention as early as possible.
Disclaimer
While AI shows significant promise in improving congenital heart defect detection rates, it is not intended to replace trained medical professionals. AI-based tools should be used as a supplementary aid to enhance clinical decision-making rather than as a standalone diagnostic method. Further research and real-world implementation studies are needed to validate these findings and ensure widespread adoption in prenatal care settings.
Provided by the Society for Maternal-Fetal Medicine.
