A new study has shown that artificial intelligence (AI) programs can improve fetal screening to detect congenital heart defects. Using a tool from medical company Bright Heart, researchers analyzed 200 second-trimester pregnancy ultrasound scans from women at 11 medical centers in two countries, including 100 scans with at least one suspicious finding. Seven obstetrician-gynecologists and seven high-risk pregnancy specialists reviewed each scan in a random order, both with and without AI assistance, looking for results that might indicate a severe heart defect. According to a report in the journal Obstetrics & Gynecology, doctors using AI detected a greater number of suspicious anomalies in less time compared to not using it. Overall, the detection rate increased from 82% to over 97%, with an 18% reduction in reading time and a 19% improvement in confidence scores. Dr. Andrei Rebarber from the Icahn School of Medicine at Mount Sinai Hospital said in a statement: "Our study should spur and encourage future research on how AI-powered programs can improve detection rates... and reduce disparity and inequality in the detection of congenital heart defects worldwide." He added: "The future of prenatal diagnostic imaging is bright when AI software is used as an aid to the physician's diagnosis."
