深度学习算法在胸部CT肋骨骨折自动检测中的评估
本研究回顾性纳入内部数据集12208例和外部数据集1613例急诊创伤患者的胸部CT,开发基于扩展U-Net架构的两个级联深度神经网络分割肋骨并检测骨折。模型在内部数据集每肋水平AUC为0.970,敏感度93.3%、特异度98.4%;结论认为该深度学习模型检测肋骨骨折可行,至少达到主治医师水平。
要点
- 深度学习模型在外部数据集检测肋骨骨折的敏感度86.2%,高于主治医师的70.5%。
- 模型特异度在内外数据集均约98.4%–98.8%,与主治医师相仿。
- 模型与人类读者一样,对不明显或模糊的病灶易漏检,需更多训练数据。
核心结论
- 所提出的深度学习模型可在胸部CT上识别肋骨骨折,至少达到主治医师水平。
研究设计与样本
- 回顾性研究,内部数据集12208例急诊创伤患者,外部数据集1613例急诊创伤患者,均行胸部CT。
- 开发两个基于扩展U-Net架构的级联深度神经网络,分别用于肋骨分割和肋骨骨折检测。
- 模型性能在内部和外部数据集上以95%CI评估,并与主治医师阅片采用t检验比较。
诊断标准与关键数值
- 内部数据集每肋水平检测骨折的AUC为0.970(95%CI: 0.968, 0.972),敏感度93.3%(95%CI: 92.0%, 94.4%),特异度98.4%(95%CI: 98.3%, 98.5%)。
- 外部数据集每肋水平检测骨折的AUC为0.943(95%CI: 0.941, 0.945),敏感度86.2%(95%CI: 85.0%, 87.3%),特异度98.8%(95%CI: 98.7%, 98.9%)。
- 主治医师在外部数据集的敏感度为70.5%(95%CI: 69.3%, 71.8%)(p < .0001),特异度为98.8%(95%CI: 98.7%, 98.9%)(p = 0.175)。
临床应用与阅片要点
- 模型可作为胸部CT肋骨骨折自动检测的辅助工具,尤其适用于急诊创伤患者。
- 需注意模型对不明显或模糊病灶的漏检可能,阅片时应结合人工复核。
局限性
- 模型与人类读者一样,对目标病灶的不明显性和模糊性敏感,细微病灶需更多训练数据才能达到可比检测性能。
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
OBJECTIVES: To evaluate deep neural networks for automatic rib fracture detection on thoracic CT scans and to compare its performance with that of attending-level radiologists using a large amount of datasets from multiple medical institutions.
METHODS: In this retrospective study, an internal dataset of 12,208 emergency room (ER) trauma patients and an external dataset of 1613 ER trauma patients taking chest CT scans were recruited. Two cascaded deep neural networks based on an extended U-Net architecture were developed to segment ribs and detect rib fractures respectively. Model performance was evaluated with a 95% confidence interval (CI) on both the internal and external dataset, and compared with attending-level radiologist readings using t test.
RESULTS: On the internal dataset, the AUC of the model for detecting fractures at per-rib level was 0.970 (95% CI: 0.968, 0.972) with sensitivity of 93.3% (95% CI: 92.0%, 94.4%) at a specificity of 98.4% (95% CI: 98.3%, 98.5%). On the external dataset, the model obtained an AUC of 0.943 (95% CI: 0.941, 0.945) with sensitivity of 86.2% (95% CI: 85.0%, 87.3%) at a specificity of 98.8% (95% CI: 98.7%, 98.9%), compared to the sensitivity of 70.5% (95% CI: 69.3%, 71.8%) (p < .0001) and specificity of 98.8% (95% CI: 98.7%, 98.9%) (p = 0.175) by attending radiologists.
CONCLUSIONS: The proposed DL model is a feasible approach to identify rib fractures on chest CT scans, at the very least, reaching a level on par with attending-level radiologists.
KEY POINTS: • Deep learning-based algorithms automatically detected rib fractures with high sensitivity and reasonable specificity on chest CT scans. • The performance of deep learning-based algorithms reached comparable diagnostic measures with attending level radiologists for rib fracture detection on chest CT scans. • The deep learning models, similar to human readers, were susceptible to the inconspicuity and ambiguity of target lesions. More training data was required for subtle lesions to achieve comparable detection performance.
引用
- Wang, Wu, Ye, et al. Assessment of automatic rib fracture detection on chest CT using a deep learning algorithm. Eur Radiol. 2022. PMID: 36214848. doi:10.1007/s00330-022-09156-w
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。