文献速递儿科影像骨肌DR

儿科骨龄评估的人为差异量化及AI评估基准

Quantifying human variability in pediatric bone age assessment: a multi-institutional multi-reader study and benchmark for AI evaluation.
PubMed 文献 · Pediatr Radiol · 2026-09-01

一项多中心多阅片者研究纳入1285例左手及腕部X线片,由22名放射科医师按Greulich-Pyle图谱评估骨龄,量化图像、阅片者及机构层面差异,并比较两种自动化方法与人类阅片者表现。结果显示低龄及男性患者人为差异更大,自动化方法性能处于或优于人类差异范围。

要点

  • 低龄儿童骨龄人为差异大于大龄儿童,报告时需注意不确定性。
  • BoneXpert较单个医师误差更小且更稳定,可作为对照参考。
  • 男童及低龄组阅片差异更大,应警惕系统性偏倚。

核心结论

  • Greulich-Pyle法骨龄评估存在可测量的人为阅片差异,尤以低龄儿童和男性患者为著。
  • 自动化方法表现处于或超出观察到的人类差异范围,可为AI性能解读提供量化基准。

研究设计与样本

  • 多阅片者多病例研究,纳入1285例左手及腕部X线片,来自5个美国学术中心。
  • 每例由4名放射科医师独立解读,阅片者来自9个机构的22名医师池。
  • 采用Greulich-Pyle图谱进行骨龄估算,使用混合效应模型评估图像、阅片者及机构层面差异。
  • 两名自动化方法分别为BoneXpert和一种深度学习算法,以三阅片者共识作为参考标准进行可互换性分析。

诊断标准与关键数值

  • 图像内放射科医师间差异:低龄患者(<12岁)标准差8.7个月,大龄患者标准差4.2个月,P<0.001。
  • 机构间差异:校正患者年龄后从8.9个月降至0.3个月。
  • 阅片者间差异在年龄校正后保持稳定,为1.2个月。
  • 男性患者和低龄组的差异更大,数名放射科医师存在与年龄或性别相关的系统性偏倚。
  • 平均绝对误差:BoneXpert为4.8个月,单个医师为6.5个月,深度学习算法为6.3个月。

临床应用与阅片要点

  • 报告低龄儿童及男童骨龄时应意识到阅片差异较大,可考虑多次复核或参考自动化工具。
  • 自动化骨龄方法可作为一致性参考,但需注意其重大偏差发生率并非为零。
  • 机构间差异主要来自患者年龄构成差异,校正后显著减小,提示标准化训练有助减少偏倚。

局限性

  • 该研究基于美国学术中心数据,结果向其他人群或机构的推广性不确定。
  • 仅使用左手及腕部X线片,其他部位或成像方式下的人为差异未评估。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

英文原文(摘要)· 点击展开

BACKGROUND: Bone age assessment using the Greulich and Pyle atlas is widely used in pediatrics but is subject to substantial reader variability. As artificial intelligence (AI) systems increasingly support clinical workflows, understanding the magnitude and structure of human variability is essential for contextualizing AI performance.

OBJECTIVE: To quantify human variability in pediatric bone age assessment across multiple institutions and to compare human variability with two automated bone age estimation methods.

MATERIALS AND METHODS: In this multi-reader, multi-case study, 1,285 left-hand and wrist radiographs from five US academic centers were independently interpreted by four radiologists per case from a centrally administered pool of 22 radiologists recruited across nine institutions. Bone age estimates were obtained using the Greulich and Pyle atlas. Variability was assessed at the image, rater, and institution levels using mixed-effects modeling. Inter-reader variability was estimated while accounting for patient age and sex. Performance of two automated methods (BoneXpert and a deep-learning algorithm) was evaluated in an interchangeability analysis using a three-rater consensus reference. Accuracy metrics included mean absolute error (MAE), root mean square error (RMSE), and rates of substantial deviation (>1.8 years).

RESULTS: Within-image variability among radiologists was higher in younger patients (<12 years) than in older patients (standard deviation (SD) 8.7 months vs. 4.2 months, P<0.001). Between-institution variability decreased substantially after adjustment for patient age (8.9 months to 0.3 months). Inter-rater variability remained stable after adjustment (1.2 months). Variability was greater in male patients and younger age groups. Several radiologists demonstrated systematic biases related to age or sex. In comparison with individual human readers, BoneXpert showed lower error (MAE 4.8 months vs. 6.5 months) and variability, while the deep-learning algorithm performed similarly to human readers (MAE 6.3 months). Substantial deviations occurred in 0.7% of BoneXpert estimates, 2.4% of deep-learning estimates, and 3.4% of single human readings.

CONCLUSION: Pediatric bone age assessment using the Greulich and Pyle method demonstrates measurable human reader variability, particularly in younger children and male patients. Automated methods performed within or beyond the observed range of human variability. These findings provide a quantitative benchmark for interpreting AI performance in skeletal maturity assessment.

来源:Pediatr Radiol,查看原文。版权归原作者所有。

引用

  1. Long, Larson, Thodberg, et al. Quantifying human variability in pediatric bone age assessment: a multi-institutional multi-reader study and benchmark for AI evaluation. Pediatr Radiol. 2026. PMID: 42678523. doi:10.1007/s00247-026-06758-0

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。