文献速递儿科影像骨肌胸部DR

基于深度学习预测2岁以下儿童正位胸片肋骨骨折的概念验证研究

Deep learning-based prediction of rib fracture presence in frontal radiographs of children under two years of age: a proof-of-concept study.
PubMed 文献 · Br J Radiol · 2023-03-03

本研究为回顾性概念验证研究,纳入1311张2岁以下儿童正位胸片(其中653张有肋骨骨折),使用ResNet-50和DenseNet-121架构进行二分类。验证集上两模型AUC-ROC分别为0.89和0.88;测试集上ResNet-50的AUC-ROC为0.84,敏感度81%、特异度70%;DenseNet-50的AUC为0.82,敏感度72%、特异度79%。结论认为深度学习可自动检测幼儿胸片肋骨骨折,性能与儿科放射科医师相当。

要点

  • 深度学习模型在2岁以下儿童胸片上检测肋骨骨折的AUC-ROC达0.82–0.89。
  • 测试集敏感度72%–81%、特异度70%–79%,性能与儿科放射科医师相当。
  • 研究提示该技术可用于疑似虐待或非意外创伤儿童的骨折筛查辅助。

关键影像表现/征象

  • 梯度加权类激活映射用于突出与深度学习模型预测最相关的区域。

诊断标准与关键数值

  • 验证集:ResNet-50的AUC-ROC为0.89,DenseNet-121的AUC-ROC为0.88。
  • 测试集:ResNet-50的AUC-ROC为0.84,敏感度81%,特异度70%。
  • 测试集:DenseNet-50的AUC为0.82,敏感度72%,特异度79%。

临床应用与阅片要点

  • 对2岁以下儿童胸片,可参考深度学习模型输出辅助判断肋骨骨折,尤其疑似虐待或非意外创伤时。
  • 需注意模型在测试集敏感度72%–81%、特异度70%–79%,存在漏诊和误诊可能。

局限性

  • 原文指出需在大型多机构数据集上进一步评估以确定泛化性。
  • 本研究为单中心回顾性概念验证研究。

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

OBJECTIVE: In this proof-of-concept study, we aimed to develop deep-learning-based classifiers to identify rib fractures on frontal chest radiographs in children under 2 years of age.

METHODS: This retrospective study included 1311 frontal chest radiographs (radiographs with rib fractures, n = 653) from 1231 unique patients (median age: 4 m). Patients with more than one radiograph were included only in the training set. A binary classification was performed to identify the presence or absence of rib fractures using transfer learning and Resnet-50 and DenseNet-121 architectures. The area under the receiver operating characteristic curve (AUC-ROC) was reported. Gradient-weighted class activation mapping was used to highlight the region most relevant to the deep learning models' predictions.

RESULTS: On the validation set, the ResNet-50 and DenseNet-121 models obtained an AUC-ROC of 0.89 and 0.88, respectively. On the test set, the ResNet-50 model demonstrated an AUC-ROC of 0.84 with a sensitivity of 81% and specificity of 70%. The DenseNet-50 model obtained an AUC of 0.82 with 72% sensitivity and 79% specificity.

CONCLUSION: In this proof-of-concept study, a deep learning-based approach enabled the automatic detection of rib fractures in chest radiographs of young children with performances comparable to pediatric radiologists. Further evaluation of this approach on large multi-institutional data sets is needed to assess the generalizability of our results.

ADVANCES IN KNOWLEDGE: In this proof-of-concept study, a deep learning-based approach performed well in identifying chest radiographs with rib fractures. These findings provide further impetus to develop deep learning algorithms for identifying rib fractures in children, especially those with suspected physical abuse or non-accidental trauma.

来源:Br J Radiol,查看原文。版权归原作者所有。

引用

  1. Ghosh, Bose, Patton, et al. Deep learning-based prediction of rib fracture presence in frontal radiographs of children under two years of age: a proof-of-concept study. Br J Radiol. 2023. PMID: 36802807. doi:10.1259/bjr.20220778

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。