人工智能在骨折检测中的应用:基于深度卷积神经网络的迁移学习
研究将非医学图像预训练的Inception v3网络顶层用腕关节侧位X线片重新训练,构建骨折与无骨折二分类模型,训练集经八倍数据增强后共11112幅图像,另用100幅腕关节X线片做最终测试。测试AUC为0.954,取兼顾敏感度与特异度的诊断界值时敏感度为0.9、特异度为0.88。作者认为该AUC与现有先进水平相当,为卷积神经网络迁移学习用于平片骨折检测提供了概念验证。
要点
- 迁移学习模型在100幅腕关节X线片测试中AUC为0.954。
- 取兼顾敏感度与特异度的界值时,敏感度0.9、特异度0.88。
- 作者认为该技术 largely transferable,在医学影像中有许多潜在应用。
核心结论
- 该测试的AUC与现有先进水平相当,为从卷积神经网络迁移学习用于平片骨折检测提供了概念验证。
- 该技术在很大程度上可迁移,因此在医学影像中有许多潜在应用,可能带来工作流程效率和临床风险降低的显著改善。
研究设计与样本
- 研究将Inception v3网络的顶层用腕关节侧位X线片重新训练,生成将新检查分类为「骨折」或「无骨折」的模型。
- 模型训练使用来自初始1389幅X线片(695幅「骨折」和694幅「无骨折」)经八倍数据增强技术后的共11112幅图像。
- 训练数据集按80:10:10分为训练组、验证组和测试组。
- 另用100幅腕关节X线片(包括50幅「骨折」和50幅「无骨折」图像)进行最终测试和统计分析。
诊断标准与关键数值
- 该测试的受试者工作特征曲线下面积(AUC)为0.954。
- 将诊断界值设在兼顾敏感度与特异度最大化的阈值时,敏感度为0.9、特异度为0.88。
临床应用与阅片要点
- 作者认为该技术在很大程度上可迁移,在医学影像中有许多潜在应用,可能带来工作流程效率和临床风险降低的显著改善。
- 该研究仅使用中等样本量即达到上述AUC,作者以此作为迁移学习用于平片骨折检测的概念验证。
局限性
- 作者指出该研究仅使用中等样本量即取得上述结果,并以此作为概念验证。
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
AIM: To identify the extent to which transfer learning from deep convolutional neural networks (CNNs), pre-trained on non-medical images, can be used for automated fracture detection on plain radiographs.
MATERIALS AND METHODS: The top layer of the Inception v3 network was re-trained using lateral wrist radiographs to produce a model for the classification of new studies as either "fracture" or "no fracture". The model was trained on a total of 11,112 images, after an eightfold data augmentation technique, from an initial set of 1,389 radiographs (695 "fracture" and 694 "no fracture"). The training data set was split 80:10:10 into training, validation, and test groups, respectively. An additional 100 wrist radiographs, comprising 50 "fracture" and 50 "no fracture" images, were used for final testing and statistical analysis.
RESULTS: The area under the receiver operator characteristic curve (AUC) for this test was 0.954. Setting the diagnostic cut-off at a threshold designed to maximise both sensitivity and specificity resulted in values of 0.9 and 0.88, respectively.
CONCLUSION: The AUC scores for this test were comparable to state-of-the-art providing proof of concept for transfer learning from CNNs in fracture detection on plain radiographs. This was achieved using only a moderate sample size. This technique is largely transferable, and therefore, has many potential applications in medical imaging, which may lead to significant improvements in workflow productivity and in clinical risk reduction.
引用
- Kim DH, MacKinnon T. Artificial intelligence in fracture detection: transfer learning from deep convolutional neural networks. Clin Radiol. 2017. PMID: 29269036. doi:10.1016/j.crad.2017.11.015
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。