卷积神经网络算法在常规CT扫描中机会性检测椎体骨折的外部验证
在年龄≥50岁的中国患者中回顾性选取胸腹部及胸腰椎CT扫描,评估既往训练的卷积神经网络模型自动检测椎体骨折的性能。共纳入4810例受试者,628例(13.1%)扫描存在≥1个2-3级椎体骨折;模型识别≥1个中重度骨折扫描的AUROC为0.94(95%CI:0.93-0.95),敏感度94%(95%CI:92-96%),特异度93%(95%CI:93-94%)。结论认为该算法在胸腹部CT扫描中识别椎体骨折表现优异。
要点
- 模型识别含≥1个中重度椎体骨折CT扫描的AUROC为0.94(95%CI:0.93-0.95),敏感度94%,特异度93%。
- 4810例受试者中628例扫描(13.1%)存在≥1个2-3级椎体骨折,共899个骨折椎体占48584个可视椎体的1.9%。
- 外部验证结果提示该算法可为放射科医生在常规胸腹部CT中早期识别椎体骨折提供机会。
核心结论
- 该算法在年龄≥50岁中国患者胸腹部CT扫描队列中识别椎体骨折表现优异。
研究设计与样本
- 研究利用两项中国研究和临床数据回顾性选取年龄≥50岁男性和女性的胸部、腹部及胸腰椎CT扫描。
- CT扫描采用Genant半定量分类评估现患椎体骨折,评估过程对临床信息设盲。
- 以参考标准阅片为对照,通过AUROC、准确率、Cohen kappa、敏感度和特异度评估卷积神经网络模型性能。
- 共纳入4810例受试者,中位年龄62岁(IQR 56-67),其中女性2654例(55.2%)。
- 扫描获取时间为2013年1月至2019年1月,使用来自3家不同制造商的16台不同CT扫描仪。
主要发现
- 4810例受试者中2773例(57.7%)为腹部CT。
- 628例扫描(13.1%)存在≥1个2-3级椎体骨折,共899个骨折椎体,占48584个可视椎体的1.9%。
- 模型识别含≥1个中重度骨折扫描的AUROC为0.94(95%CI:0.93-0.95)。
- 模型识别含≥1个中重度骨折扫描的准确率为93%(95%CI:93%-94%),kappa为0.75(95%CI:0.72-0.77)。
- 模型识别含≥1个中重度骨折扫描的敏感度为94%(95%CI:92-96%),特异度为93%(95%CI:93-94%)。
关键数值与推荐
- 椎体骨折采用Genant半定量分类评估,中重度骨折定义为2-3级。
- 模型性能以AUROC、准确率、Cohen kappa、敏感度和特异度对照参考标准阅片评价。
临床怎么用
- 外部验证结果提示该算法可为放射科医生在常规腹部和胸部CT扫描中早期识别椎体骨折提供机会。
局限性
- 原文未报告局限性。
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)。
原文
UNLABELLED: The Convolutional Neural Network algorithm achieved a sensitivity of 94% and specificity of 93% in identifying scans with vertebral fractures (VFs). The external validation results suggest that the algorithm provides an opportunity to aid radiologists with the early identification of VFs in routine CT scans of abdomen and chest. PURPOSE: To evaluate the performance of a previously trained Convolutional Neural Network (CNN) model to automatically detect vertebral fractures (VFs) in CT scans in an external validation cohort. METHODS: Two Chinese studies and clinical data were used to retrospectively select CT scans of the chest, abdomen and thoracolumbar spine in men and women aged ≥50 years. The CT scans were assessed using the semiquantitative (SQ) Genant classification for prevalent VFs in a process blinded to clinical information. The performance of the CNN model was evaluated against reference standard readings by the area under the receiver operating characteristics curve (AUROC), accuracy, Cohen's kappa, sensitivity, and specificity. RESULTS: A total of 4,810 subjects were included, with a median age of 62 years (IQR 56-67), of which 2,654 (55.2%) were females. The scans were acquired between January 2013 and January 2019 on 16 different CT scanners from three different manufacturers. 2,773 (57.7%) were abdominal CTs. A total of 628 scans (13.1%) had ≥1 VF (grade 2-3), representing 899 fractured vertebrae out of a total of 48,584 (1.9%) visualized vertebral bodies. The CNN's performance in identifying scans with ≥1 moderate or severe fractures achieved an AUROC of 0.94 (95% CI: 0.93-0.95), accuracy of 93% (95% CI: 93%-94%), kappa of 0.75 (95% CI: 0.72-0.77), a sensitivity of 94% (95% CI: 92-96%) and a specificity of 93% (95% CI: 93-94%). CONCLUSION: The algorithm demonstrated excellent performance in the identification of vertebral fractures in a cohort of chest and abdominal CT scans of Chinese patients ≥50 years.
引用
- Nicolaes J, Liu Y, Zhao Y, et al. External validation of a convolutional neural network algorithm for opportunistically detecting vertebral fractures in routine CT scans. Osteoporos Int. 2023. PMID: 37674097. doi:10.1007/s00198-023-06903-7
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。