A recent study led by Mount Sinai West doctors found AI technology helped identify more than 97 percent of serious congenital heart defects in babies
This initiative was led bymedics in the Raquel and Jaime Gilinski Department of Obstetrics, Gynecology and Reproductive Science at Mount Sinai and it centres on implementing an artificial intelligence tool that enhances ultrasounds on a large scale—resulting in earlier detection and better care for babies and families.
Congenital heart defects, or conditions present at birth that affect the heart structure, are one of the most common birth abnormalities. About 1 in 500 newborns is classified as having a severe congenital heart defect that requires urgent medical or surgical intervention for survival, according to the U.S. National Institutes of Health.
Congenital heart defects are the most common birth defect. They are present in some 48.9 million people globally.
Consequently, the Carnegie Imaging for Women, a modern OB/GYN imaging facility, becomes the first centre in New York City to use a U.S. Food and Drug Administration-approved AI software tool.
The technology was developed by the medical company BrightHeart, with the objective to make ultrasounds more accurate and efficient.
The experimental successes of trialling the new technology appear in a recent Obstetrics & Gynecology study led by Mount Sinai West doctors (“Use of Artificial Intelligence–Based Software to Aid in the Identification of Ultrasound Findings Associated With Fetal Congenital Heart Defects”). Here, the researchers used the AI technology to improve their detection rates of ultrasound findings suspicious for major congenital heart defects to more than 97 percent, with an 18 percent reduction in reading time and 19 percent improvement in confidence score.
The AI assistance in prenatal diagnosis was found to offer not only improved detection, but it also has the potential to offer significant improvement in workflow and efficiency benefits.
The researchers examined a dataset of 200 deidentified foetal ultrasound examinations between 18 and 24 weeks of gestation from 11 medical centres across two countries, including 100 with at least one suspicious finding. The study aimed to evaluate the association between the use of AI-based software and reader performance in identifying second-trimester ultrasound examinations suspicious for severe congenital heart defects.
Seven obstetrician-gynaecologists and seven maternal-foetal medicine specialists (experts in high-risk pregnancies) reviewed each examination in randomized order, both with and without AI assistance, and assessed the presence or absence of each finding suspicious for congenital heart defects with confidence scores.
The AI-assisted interpretation was associated with improved detection of lesions suspicious for severe congenital heart defects. The study demonstrated the ability of AI-based software to improve the detection of these suspicious findings via prenatal ultrasonography, as well as the overall confidence and time efficiency in interpreting these scans.
