VP17.01: Exploring a new paradigm for the fetal anomaly ultrasound scan: artificial intelligence in real‐time

Abstract

Ultrasound (US) is characterised by high levels of operator subjectivity and variability. Advances in artificial intelligence (AI) have demonstrated the potential to reduce both. This study pilots the end-to-end automation of multiple elements of the mid-trimester obstetric screening US scan using AI-enabled tools. A single centre, prospective method comparison study was conducted. Participants had both standard manual and AI-assisted US scans, each performed alternately and independently by 2 blinded sonographers. The AI tools automated the acquisition of standard plane images, measurements, and the production of a written clinical report with saved images available for review. A feedback survey captured the sonographers' perceptions of scanning. 23 subjects were studied. The average time saving per scan was 7.62min (34.7%) when using the AI-assisted method (p < 0.0001) with no difference in reporting time. There were no clinically significant differences in biometric measurements between methods. The AI tools saved a satisfactory view in 93% of the cases when considering the four core views only, and 73% for the full 13 views, compared to 98% for both using the manual scan. Survey responses suggest the AI tools helped sonographers to concentrate on image interpretation by removing disruptive recording and measurement tasks. Using AI to automate tasks during the ultrasound examination changes workflow. Separating the process of freehand scanning from image capture and measurement resulted in a faster scan. Reducing the need for sonographers to focus on repetitive tasks may allow more attention to be directed towards identification of atypical fetal anatomy. Further work is required to improve the performance of the image plane detection algorithm for use in real-time. In the future, high-quality AI tools could allow the sonographer to increase their focus on anatomical assessment for congenital anomaly detection and provide higher-quality parent-centred care.

Ähnliche Arbeiten