基于卷积神经网络机器学习方法利用乳腺X线图像数据区分非典型导管增生与导管原位癌的准确性
本研究利用149例患者共298张乳腺X线放大图像,构建15层隐藏层的卷积神经网络,区分经立体定向活检诊断的纯非典型导管增生(未升级)与导管原位癌。测试集AUC为0.86(95%CI ±0.03),汇总敏感度84.6%(95%CI ±4.0%)、特异度88.2%(95%CI ±3.0%)、诊断准确率86.7%(95%CI ±2.9)。作者认为用乳腺X线图像区分两者可行,更大数据集可能进一步改善模型。
要点
- 测试集AUC 0.86(95%CI ±0.03),敏感度84.6%、特异度88.2%(95%CI分别为±4.0%、±3.0%)。
- 输入为钙化的头尾位与内外侧斜位或外内侧位两张标准放大视图,图像缩放至128×128像素。
- 网络含15层隐藏层、5个残差层,每次卷积后dropout为0.25。
核心结论
- 将卷积神经网络用于乳腺X线图像区分纯非典型导管增生与导管原位癌是可行的,更大数据集可能进一步改善预测模型。
研究设计与样本
- 研究论文;共149例患者298张唯一图像,其中非典型导管增生组67例134张(立体定向活检诊断钙化、手术切除时未升级为导管原位癌或浸润癌),导管原位癌组82例164张;按患者随机分为训练验证集(80%患者)与测试集(20%患者);使用开源软件分割钙化并缩放至128×128像素边界框,采用15层隐藏层、5个残差层、每次卷积后dropout 0.25的卷积神经网络。
诊断标准与关键数值
- 测试集AUC为0.86(95%CI ±0.03);汇总敏感度84.6%(95%CI ±4.0%)、汇总特异度88.2%(95%CI ±3.0%)、诊断准确率86.7%(95%CI ±2.9)。
临床应用与阅片要点
- 作者认为卷积神经网络用于乳腺X线图像区分纯非典型导管增生与导管原位癌是可行的。
- 作者指出更大数据集可能进一步改善预测模型。
局限性
- 作者指出更大数据集可能进一步改善预测模型。
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. The purpose of this study was to test the hypothesis that convolutional neural networks can be used to predict which patients with pure atypical ductal hyperplasia (ADH) may be safely monitored rather than undergo surgery. MATERIALS AND METHODS. A total of 298 unique images from 149 patients were used for our convolutional neural network algorithm. A total of 134 images from 67 patients with ADH that had been diagnosed by stereotactic-guided biopsy of calcifications but had not been upgraded to ductal carcinoma in situ or invasive cancer at the time of surgical excision. A total of 164 images from 82 patients with mammographic calcifications indicated that ductal carcinoma in situ was the final diagnosis. Two standard mammographic magnification views of the calcifications (a craniocaudal view and a mediolateral or lateromedial view) were used for analysis. Calcifications were segmented using an open-source software platform and images were resized to fit a bounding box of 128 × 128 pixels. A topology with 15 hidden layers was used to implement the convolutional neural network. The network architecture contained five residual layers and dropout of 0.25 after each convolution. Patients were randomly separated into a training-and-validation set (80% of patients) and a test set (20% of patients). Code was implemented using open-source software on a workstation with an open-source operating system and a graphics card. RESULTS. The AUC value was 0.86 (95% CI, ± 0.03) for the test set. Aggregate sensitivity and specificity were 84.6% (95% CI, ± 4.0%) and 88.2% (95% CI, ± 3.0%), respectively. Diagnostic accuracy was 86.7% (95% CI, ± 2.9). CONCLUSION. It is feasible to apply convolutional neural networks to distinguish pure atypical ductal hyperplasia from ductal carcinoma in situ with the use of mammographic images. A larger dataset will likely result in further improvement of our prediction model.
引用
- Ha R, Mutasa S, Sant EPV, et al. Accuracy of Distinguishing Atypical Ductal Hyperplasia From Ductal Carcinoma In Situ With Convolutional Neural Network-Based Machine Learning Approach Using Mammographic Image Data. AJR Am J Roentgenol. 2019. PMID: 30860901. doi:10.2214/AJR.18.20250
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。