Development of an Artificial Intelligence Algorithm to Recognize Abnormal Findings at Routine Fetal Brain Ultrasound AIRFRAME: Artificial Intelligence for Recognition of Fetal Brain Anomalies

What this Trial Is About

This research aims to develop and validate an artificial intelligence (AI) software to recognize fetal brain structures and distinguish between normal and abnormal brain anatomy during the second trimester ultrasound scan. The study focuses on fetal brain abnormalities, which are challenging to diagnose prenatally. AI technology may improve detection, reduce variability between operators, shorten examination time, and optimize healthcare resources. The study is conducted by multiple fetal medicine centers and targets pregnant women undergoing routine screening for fetal anomalies. The study consists of two phases: a retrospective phase and a prospective phase. In the first phase, researchers collect and analyze ultrasound images taken between 19 and 22 weeks of gestation from various centers to develop and train the AI algorithm with both normal and abnormal fetal brain images. The second phase prospectively tests the AI algorithm in real clinical settings on patients from the participating centers to validate its performance in assessing fetal brain anatomy. Participants are singleton pregnant women between 19 and 22 weeks of pregnancy who undergo ultrasound scans. Researchers will collect clinical, ultrasound, prenatal, and postnatal data, anonymize images, and store them securely for analysis. The primary outcome is validating the AI algorithm over two years. Secondary outcomes include improving reproducibility and reducing examination time within one year. The study involves continuous monitoring and assessment of ultrasound images to support routine fetal brain screening.

AIRFRAME: Artificial Intelligence for Recognition of Fetal bRain AnoMaliEs at Second Trimester Fetal Brain Scan