Deep Neural Networks Improve Radiologists Performance in Breast Cancer Screening
Deep Neural Networks Improve Radiologists Performance in Breast Cancer Screening
复制标题
DOI：
10.1109/tmi.2019.2945514
复制
发表时间：
2020-04-01
影响因子：
10.6
通讯作者：
Geras, Krzysztof J.
中科院分区：
文献类型：
--
作者：
Wu, Nan;Phang, Jason;Geras, Krzysztof J.
我们提出了一个用于乳腺癌筛查检查分类的深度卷积神经网络，对超过200000次检查（超过1000000张图像）进行了训练和评估。当在筛查人群中进行测试时，我们的网络在预测乳腺癌的存在方面达到了0.895的AUC。我们把这种高精度归功于一些技术进步。1)我们的网络2019;的新颖的两阶段架构和训练过程，使我们能够使用高容量的补丁级网络从像素级标签学习，同时使用从宏观乳房级标签学习的网络。2)一个基于ResNet的自定义网络用作我们模型的构建块，其深度和宽度的平衡针对高分辨率医学图像进行了优化。3)对网络进行筛选BI-RADS分类的预训练，这是一项具有更多噪声标签的相关任务。4)在许多可能的选择中以最佳方式组合多个输入视图。为了验证我们的模型，我们进行了一项读者研究，有14名读者，每个阅读720筛查乳房X线检查，并表明我们的模型是准确的经验丰富的放射科医生时，提供相同的数据。我们还表明，一个混合模型，平均预测的恶性肿瘤的概率，由放射科医生与我们的神经网络的预测，是更准确的比两者分开。为了进一步了解我们的结果，我们对我们的网络2019;在筛选人群的不同亚群上的性能，模型2019;的设计，训练过程，错误及其内部表示的属性进行了彻底的分析。我们的最佳模型可在https://github.com/nyukat/breast_cancer_classifier上公开获取。
We present a deep convolutional neural network for breast cancer screening exam classification, trained, and evaluated on over 200000 exams (over 1000000 images). Our network achieves an AUC of 0.895 in predicting the presence of cancer in the breast, when tested on the screening population. We attribute the high accuracy to a few technical advances. 1) Our network 2019;s novel two-stage architecture and training procedure, which allows us to use a high-capacity patch-level network to learn from pixel-level labels alongside a network learning from macroscopic breast-level labels. 2) A custom ResNet-based network used as a building block of our model, whose balance of depth and width is optimized for high-resolution medical images. 3) Pretraining the network on screening BI-RADS classification, a related task with more noisy labels. 4) Combining multiple input views in an optimal way among a number of possible choices. To validate our model, we conducted a reader study with 14 readers, each reading 720 screening mammogram exams, and show that our model is as accurate as experienced radiologists when presented with the same data. We also show that a hybrid model, averaging the probability of malignancy predicted by a radiologist with a prediction of our neural network, is more accurate than either of the two separately. To further understand our results, we conduct a thorough analysis of our network 2019;s performance on different subpopulations of the screening population, the model 2019;s design, training procedure, errors, and properties of its internal representations. Our best models are publicly available at https://github.com/nyukat/breast_cancer_classifier.