Department of Diagnostic Radiology, Tokyo Medical and Dental University, Tokyo, Japan.
Department of Diagnostic Radiology, Tokyo Medical and Dental University, Tokyo, Japan.
Magn Reson Imaging. 2021 Jan;75:1-8. doi: 10.1016/j.mri.2020.10.003. Epub 2020 Oct 10.
We aimed to evaluate deep learning approach with convolutional neural networks (CNNs) to discriminate between benign and malignant lesions on maximum intensity projections of dynamic contrast-enhanced breast magnetic resonance imaging (MRI).
We retrospectively gathered maximum intensity projections of dynamic contrast-enhanced breast MRI of 106 benign (including 22 normal) and 180 malignant cases for training and validation data. CNN models were constructed to calculate the probability of malignancy using CNN architectures (DenseNet121, DenseNet169, InceptionResNetV2, InceptionV3, NasNetMobile, and Xception) with 500 epochs and analyzed that of 25 benign (including 12 normal) and 47 malignant cases for test data. Two human readers also interpreted these test data and scored the probability of malignancy for each case using Breast Imaging Reporting and Data System. Sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUC) were calculated.
The CNN models showed a mean AUC of 0.830 (range, 0.750-0.895). The best model was InceptionResNetV2. This model, Reader 1, and Reader 2 had sensitivities of 74.5%, 72.3%, and 78.7%; specificities of 96.0%, 88.0%, and 80.0%; and AUCs of 0.895, 0.823, and 0.849, respectively. No significant difference arose between the CNN models and human readers (p > 0.125).
Our CNN models showed comparable diagnostic performance in differentiating between benign and malignant lesions to human readers on maximum intensity projection of dynamic contrast-enhanced breast MRI.
我们旨在评估基于卷积神经网络（CNN）的深度学习方法，以区分动态对比增强乳腺磁共振成像（MRI）最大强度投影上的良性和恶性病变。
我们回顾性收集了 106 例良性（包括 22 例正常）和 180 例恶性病例的动态对比增强乳腺 MRI 的最大强度投影数据，用于训练和验证数据。构建了 CNN 模型，使用 500 个时期的 CNN 架构（DenseNet121、DenseNet169、InceptionResNetV2、InceptionV3、NasNetMobile 和 Xception）计算恶性肿瘤的概率，并对 25 例良性（包括 12 例正常）和 47 例恶性病例的测试数据进行了分析。两名人类读者也对这些测试数据进行了解读，并使用乳腺成像报告和数据系统（Breast Imaging Reporting and Data System）为每个病例评分恶性肿瘤的概率。计算了敏感性、特异性、准确性和受试者工作特征曲线（ROC）下的面积（AUC）。
CNN 模型的平均 AUC 为 0.830（范围，0.750-0.895）。最好的模型是 InceptionResNetV2。该模型、Reader 1 和 Reader 2 的敏感性分别为 74.5%、72.3%和 78.7%；特异性分别为 96.0%、88.0%和 80.0%；AUC 分别为 0.895、0.823 和 0.849。CNN 模型与人类读者之间无显著差异（p>0.125）。
在动态对比增强乳腺 MRI 的最大强度投影上，我们的 CNN 模型在区分良性和恶性病变方面的表现与人类读者相当。
