Accuracy of Distinguishing Atypical Ductal Hyperplasia From Ductal Carcinoma In Situ With Convolutional Neural Network-Based Machine Learning Approach Using Mammographic Image Data.
Accuracy of Distinguishing Atypical Ductal Hyperplasia From Ductal Carcinoma In Situ With Convolutional Neural Network-Based Machine Learning Approach Using Mammographic Image Data.
复制标题
使用乳房X线图像数据区分卷积神经网络的机器学习方法，将非典型导管增生与导管癌区分开的精度。
DOI：
10.2214/ajr.18.20250
复制
发表时间：
2019-05
期刊：
影响因子：
--
通讯作者：
Jambawalikar S
中科院分区：
文献类型：
--
作者：
Ha R;Mutasa S;Sant EPV;Karcich J;Chin C;Liu MZ;Jambawalikar S
本研究的目的是检验卷积神经网络可用于预测哪些纯非典型导管增生（ADH）患者可以安全监测而不是接受手术的假设。我们的卷积神经网络算法使用了来自149名患者的298张独特图像。来自67例ADH患者的共134张图像，这些患者通过立体定向引导钙化活检诊断，但在手术切除时未升级为导管原位癌或浸润性癌。82例患者的164张乳腺摄影钙化图像表明导管原位癌是最终诊断。使用钙化的两个标准乳房X线摄影放大视图（头尾视图和内外侧或内侧视图）进行分析。使用开源软件平台分割钙化，并调整图像大小以适应128 × 128像素的边界框。使用具有15个隐藏层的拓扑结构来实现卷积神经网络。该网络架构包含五个残留层，每次卷积后的dropout为0.25。患者被随机分为训练和验证集（80%的患者）和测试集（20%的患者）。代码是在一个带有开源操作系统和图形卡的工作站上使用开源软件实现的。试验集的AUC值为0.86（95% CI，± 0.03）。总体敏感性和特异性分别为84.6%（95% CI，± 4.0%）和88.2%（95% CI，± 3.0%）。诊断准确率为86.7%（95% CI，± 2.9）。应用卷积神经网络在乳腺摄影图像上区分单纯性导管不典型增生和导管原位癌是可行的。更大的数据集可能会进一步改进我们的预测模型。
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. 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. 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). 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.