Prior to Initiation of Chemotherapy, Can We Predict Breast Tumor Response? Deep Learning Convolutional Neural Networks Approach Using a Breast MRI Tumor Dataset
Prior to Initiation of Chemotherapy, Can We Predict Breast Tumor Response? Deep Learning Convolutional Neural Networks Approach Using a Breast MRI Tumor Dataset
复制标题
DOI：
10.1007/s10278-018-0144-1
复制
发表时间：
2019-10-01
影响因子：
4.4
通讯作者：
Jambawalikar, Sachin
中科院分区：
文献类型：
--
作者：
Ha, Richard;Chin, Christine;Jambawalikar, Sachin
我们假设卷积神经网络（CNN）可以用于在开始化疗之前使用乳腺MRI肿瘤数据集预测新辅助化疗（NAC）反应。一项机构审查委员会批准的对2009年1月至2016年6月我们数据库的回顾性审查确定了141例局部晚期乳腺癌患者，这些患者（1）在开始NAC前接受了乳腺MRI，（2）成功完成了基于阿霉素/紫杉烷的NAC，（3）接受了手术切除，并提供了最终手术病理学数据。根据最终手术病理学证实的NAC缓解，将患者分为三组：完全缓解（第1组）、部分缓解（第2组）和无缓解/进展（第3组）。共对141个肿瘤的3107个体积切片进行了评价。在第一个T1增强后动态图像上识别乳腺肿瘤，并进行3D分割。CNN由10个卷积层，4个最大池化层组成，在完全连接层之后的丢弃率为50%。实施了Dropout、增强和L2正则化以防止数据的过度拟合。非线性函数由校正线性单元（ReLU）建模。在卷积层和ReLU层之间使用批量归一化，以限制训练期间层激活的漂移。评价了三级新辅助治疗预测模型（第1组、第2组或第3组）。CNN在新辅助治疗反应的三级预测中实现了88%的总体准确性。分析了区分一组和另两组的三级预测。第1组的特异性为95.1% ± 3.1%，敏感性为73.9% ± 4.5%，准确性为87.7% ± 0.6%。第2组（部分缓解）的特异性为91.6% ± 1.3%，敏感性为82.4% ± 2.7%，准确性为87.7% ± 0.6%。第3组（无缓解/进展）的特异性为93.4% ± 2.9%，敏感性为76.8% ± 5.7%，准确性为87.8% ± 0.6%。目前的深度CNN架构可以使用在开始化疗之前获得的乳腺MRI数据集进行训练，以预测NAC治疗反应。更大的数据集可能会改善我们的预测模型。
We hypothesize that convolutional neural networks (CNN) can be used to predict neoadjuvant chemotherapy (NAC) response using a breast MRI tumor dataset prior to initiation of chemotherapy. An institutional review board-approved retrospective review of our database from January 2009 to June 2016 identified 141 locally advanced breast cancer patients who (1) underwent breast MRI prior to the initiation of NAC, (2) successfully completed adriamycin/taxane-based NAC, and (3) underwent surgical resection with available final surgical pathology data. Patients were classified into three groups based on their NAC response confirmed on final surgical pathology: complete (group 1), partial (group 2), and no response/progression (group 3). A total of 3107 volumetric slices of 141 tumors were evaluated. Breast tumor was identified on first T1 postcontrast dynamic images and underwent 3D segmentation. CNN consisted of ten convolutional layers, four max-pooling layers, and dropout of 50% after a fully connected layer. Dropout, augmentation, and L2 regularization were implemented to prevent overfitting of data. Non-linear functions were modeled by a rectified linear unit (ReLU). Batch normalization was used between the convolutional and ReLU layers to limit drift of layer activations during training. A three-class neoadjuvant prediction model was evaluated (group 1, group 2, or group 3). The CNN achieved an overall accuracy of 88% in three-class prediction of neoadjuvant treatment response. Three-class prediction discriminating one group from the other two was analyzed. Group 1 had a specificity of 95.1% +/- 3.1%, sensitivity of 73.9% +/- 4.5%, and accuracy of 87.7% +/- 0.6%. Group 2 (partial response) had a specificity of 91.6% +/- 1.3%, sensitivity of 82.4% +/- 2.7%, and accuracy of 87.7% +/- 0.6%. Group 3 (no response/progression) had a specificity of 93.4% +/- 2.9%, sensitivity of 76.8% +/- 5.7%, and accuracy of 87.8% +/- 0.6%. It is feasible for current deep CNN architectures to be trained to predict NAC treatment response using a breast MRI dataset obtained prior to initiation of chemotherapy. Larger dataset will likely improve our prediction model.