Artificial Intelligence to Assist in the Screening Fetal Anomaly Ultrasound Scan (PROMETHEUS): A Randomised Controlled Trial

Author:

Day Thomas GORCID,Matthew JacquelineORCID,Budd Samuel F,Farruggia Alfonso,Venturini Lorenzo,Wright Robert,Jamshidi Babak,To Meekai,Ling Huazen,Lai Jonathon,Tan Min Yi,Brown Matthew,Guy Gavin,Casagrandi Davide,Arechvo Anastasija,Syngelaki Argyro,Lloyd David,Zidere Vita,Vigneswaran Trisha,Miller Owen,Akolekar Ranjit,Nanda Surabhi,Nicolaides Kypros,Kainz Bernhard,Simpson John M,Hajnal Jo V,Razavi Reza

Abstract

AbstractBackgroundArtificial intelligence (AI) has shown potential in improving the performance of screening fetal anomaly ultrasound scans. We aimed to assess the effect of AI on fetal ultrasound scanning, in terms of diagnostic performance, biometry, scan duration, and sonographer cognitive load.MethodsThis was a randomised, single centre, open label trial in a large teaching hospital. Pregnant participants with fetal congenital heart disease (CHD) and with healthy fetuses were recruited and scanned with both methods. Screening sonographers were recruited from regional hospitals and were randomised to scan with the AI tool or in the standard fashion, blinded to the fetal CHD status. For the AI-assisted scans, the AI models identified and saved 13 standard image planes, and measured four biometrics.Findings78 pregnant participants (26 with fetal CHD) and 58 sonographers were recruited. The sensitivity and specificity of the AI-assisted scan in detecting fetal malformation was 88.9% and 98.0% respectively, with the standard scan achieving 81.5% and 92.2% (not significant). AI-assisted scans were significantly shorter than standard scans (median 11.4 min vs 19.7 min, p <0.001). Sonographer cognitive load was significantly lower in the AI-assisted group (median NASA TLX score 35.2 vs 46.5, p <0.001). For all biometrics, the AI repeatability and reproducibility was superior to manual measurements.InterpretationAI assistance in the routine fetal anomaly ultrasound scan results in a significant time saving, along with a reduction in sonographer cognitive load, without a reduction in diagnostic performance.FundingThe study was funded by an NIHR doctoral fellowship (NIHR301448) and was supported by grants from the Wellcome Trust (IEH Award, 102431), by core funding from the Wellcome Trust/EPSRC Centre for Medical Engineering (WT203148/Z/16/Z), and the London AI Centre for Value Based Healthcare via funding from the Office for Life Sciences.Research in contextEvidence before this studyWe undertook a systematic review exploring evidence for the use of artificial intelligence (AI) to assist in the performance of fetal anomaly ultrasound scans by automating anatomical standard plane detection, and / or automating fetal biometric measurements. We searched PubMed, the Cochrane Library, and clinicaltrials.gov databases using the following search terms: ((artificial intelligence) OR (AI) OR (machine learning) OR (neural network*) OR (deep learning)) AND ((fetal) OR (foetal) OR (fetus) OR (foetus) OR (obstetric) OR (antenatal) OR (prenatal) OR (pregnan*)) AND (ultrasound) AND ((biometr*) OR (measurement) OR (growth) OR (size) OR (plane) OR (view) OR (femur) OR (head) OR (biparietal) OR (abdom*)). We limited the results to the 10-year period September 2013 – September 2023. 770 papers were identified, of which 55, 39, and 33 were deemed relevant after title, abstract, and text screening respectively. Of the 33 papers, 14 focused on biometric measurements, 14 on plane detection, and 5 included both. Only one paper tested the AI models prospectively with real-time feedback to the sonographer, but this study did not include randomisation or fetal pathology. No randomised clinical trials comparing AI-assisted ultrasound scans with standard scans have been performed previously.Added value of this studyTo our knowledge, this is the first randomised controlled trial investigating the use of AI in fetal ultrasound screening. In this trial we assess the use of AI-assistance to automatically undertake some aspects of the scan (automatic anatomical standard plane detection and saving, and measurement of biometric parameters), and measure the effect on overall diagnostic performance, as well as scan duration, sonographer cognitive load, image quality, and repeatability and reproducibility of biometric measurements. AI assistance resulted in a significantly lower scan duration and sonographer cognitive load, whilst maintaining the quality of the scan in terms of diagnostic performance and biometric measurements.Implications of all the available evidenceThe results from this trial are encouraging and suggest that AI assistance may offer real clinical benefit to sonographers undertaking fetal ultrasound screening. The reduced scan duration means that sonographers may have more time to focus on other aspects of the scan, such as communication with parents. The automatically measured biometrics were both more repeatable and reproducible compared to manual measurements, which may improve the accuracy with which fetal growth and health can be assessed. Further studies combining this work with AI models that can directly detect fetal structural malformations will be important, to improve the overall antenatal detection of fetal anomalies.

Publisher

Cold Spring Harbor Laboratory

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3