Feasibility and reproducibility of an image-scoring method for quality control of fetal biometry in the second trimester
Feasibility and reproducibility of an image-scoring method for quality control of fetal biometry in the second trimester
复制标题
DOI：
10.1002/uog.2665
复制
发表时间：
2006-01-01
影响因子：
7.1
通讯作者：
Ville, Y
中科院分区：
文献类型：
--
作者：
Salomon, LJ;Bernard, JP;Ville, Y
目标 胎儿超声培训计划和认证流程的需求已变得显而易见。本研究的目的是评估基于评分的妊娠中期胎儿生物测量质量控制系统的可行性。方法由四名操作员使用同一台超声仪对 20-24 周时的双顶径和珠周、腹围和股骨长度进行标准测量。从每位操作员的超声数据库中任意选择 25 张带有卡尺的头颅、腹部和股骨图像，并进行匿名处理。这 300 张图像由三位经验丰富的评审员进行了分析，他们对操作员的身份一无所知。首先对每张图像进行主观评估，然后根据腹部和头部测量的六个标准以及股骨长度的四个标准进行评分，腹部和头部生物测量为六分，股骨长度为四分。对于主观评价，使用百分比一致性和调整后的卡帕值来分析审稿人之间的差异。对于客观评价，评审者之间的评分差异一分或更少被认为是良好的一致性。使用任意选择的每种检查类型的 40 张图像来评估评审者内部的变异性。结果 评审者之间的分数分布相似。一名操作员获得了明显较低的分数，而其他三名操作员的分数却很差，但结果相当好。每个评审者给出的平均分数没有统计差异，并且 84-90% 的案例一致性良好。在 90-100% 的情况下，审稿人内部的一致性良好，每个审稿人的分数相似。结论 基于图像评分的质量控制策略是可行的，并且可以实现公平到良好的审稿人之间和内部的再现性。这种方法对评估常规超声检查质量的潜在贡献应该进行更大规模的测试。版权所有 (c) 2005 ISUOG。由约翰·威利父子有限公司出版
Objectives The need for training programs and certification processes in fetal ultrasound has become obvious. The Purpose of this study was to evaluate the feasibility of a score-based quality control system for fetal biometry in the second trimester.Methods Standard measurements of biparietal diameter and bead circumference, abdominal circumference, and femur length at 20-24 weeks bad been made by four operators using the same ultrasound machine. Twenty-five of each of the cephalic, abdominal and femoral images with the calipers in place were selected arbitrarily from each operator's ultrasound database and anonymized. These 300 images were analyzed by three experienced reviewers blinded to the operator's identity. Each image was first evaluated subjectively and then scored according to six criteria for abdominal and cephalic measurements and four criteria for femur length making a six-point score for abdominal and cephalic biometry and a four-point score for femur length. For subjective evaluation, inter-reviewer differences were analyzed using percentage agreement and adjusted kappa. For objective evaluation, a difference in scoring of one point or less among reviewers was considered good agreement. Intrareviewer variability was assessed using 40 images of each type of examination selected arbitrarily.Results The distribution of scores was similar between reviewers. One operator obtained significantly lower scores whereas the other three bad good and comparable results. There was no statistical difference in the mean score attributed by each reviewer and agreement was good in 84-90% of the cases. Intrareviewer agreement was good in 90-100% of the cases, with similar scores for each reviewer.Conclusion A quality control policy based on image scoring is feasible and allows for fair to good inter- and intrareviewer reproducibility. The potential contribution of this approach to assess the quality of routine ultrasound examinations should be tested on a larger scale. Copyright (c) 2005 ISUOG. Published by John Wiley & Sons, Ltd.