Using Artificial Intelligence to Improve Quality Control of Early Pregnancy Ultrasound Images Focusing on Key Fetal Sections to Reduce Misdiagnosis of Conditions

What This Trial Is About

Researchers are exploring the use of artificial intelligence to improve quality control of early pregnancy ultrasound images, focusing on key fetal sections like the median sagittal, nuchal translucency NT, and choroid plexus views. This study collaborates with leading medical centers to collect extensive ultrasound data to develop a deep learning model. The goal is to assist clinicians by ensuring ultrasound images meet quality standards, potentially reducing missed or incorrect diagnoses of fetal conditions such as Down Syndrome and neural system deformities. The study gathers clinical information and ultrasound images from early pregnant women at multiple hospitals. The deep learning model will identify important anatomical areas in the ultrasound scans and assess whether the images meet established quality criteria. This process supports real-time quality assessment during ultrasound exams, guiding clinicians to standardize their imaging techniques and improve diagnostic accuracy. Participants provide ultrasound images that clearly show the specified fetal sections along with detailed personal information. Researchers will analyze these images to evaluate the accuracy and performance of the AI quality control tool, using measures such as the precision-recall curve to assess outcomes within one month. The study involves monitoring image quality and clinical data, with the total participation duration varying based on image collection and analysis timelines.

Quality Control of Ultrasound Images During Early Pregnancy Via AI