Brain tumor segmentation with Deep Neural Networks
Brain tumor segmentation with Deep Neural Networks
复制标题
DOI：
10.1016/j.media.2016.05.004
复制
发表时间：
2017-01-01
影响因子：
10.9
通讯作者：
Larochelle, Hugo
中科院分区：
文献类型：
--
作者：
Havaei, Mohammad;Davy, Axel;Larochelle, Hugo
本文提出了一种基于深度神经网络（dnn）的全自动脑肿瘤分割方法。所提出的网络是针对MR图像中的胶质母细胞瘤（包括低级别和高级别）量身定制的。就其本质而言，这些肿瘤可以出现在大脑的任何地方，几乎有任何形状、大小和对比度。这些原因促使我们探索一种机器学习解决方案，该解决方案利用灵活、高容量的深度神经网络，同时非常高效。在这里，我们给出了不同的模型选择的描述，我们已经发现是必要的，以获得有竞争力的表现。我们特别探索了基于卷积神经网络（CNN）的不同架构，即专门适应图像数据的dnn。我们提出了一种不同于传统计算机视觉的CNN架构。我们的CNN同时利用了本地特征和更多的全球背景特征。此外，与cnn的大多数传统用途不同，我们的网络使用的最后一层是一个全连接层的卷积实现，这使得速度提高了40倍。我们还描述了一个两阶段的训练程序，使我们能够解决与肿瘤标签不平衡相关的困难。最后，我们探索了一种级联架构，其中基本CNN的输出被视为后续CNN的附加信息源。2013年BRATS测试数据集报告的结果显示，我们的架构比目前公布的最先进的技术改进了30多倍。(C) 2016 Elsevier B.V.版权所有
In this paper, we present a fully automatic brain tumor segmentation method based on Deep Neural Networks (DNNs). The proposed networks are tailored to glioblastomas (both low and high grade) pictured in MR images. By their very nature, these tumors can appear anywhere in the brain and have almost any kind of shape, size, and contrast. These reasons motivate our exploration of a machine learning solution that exploits a flexible, high capacity DNN while being extremely efficient. Here, we give a description of different model choices that we've found to be necessary for obtaining competitive performance. We explore in particular different architectures based on Convolutional Neural Networks (CNN), i.e. DNNs specifically adapted to image data.We present a novel CNN architecture which differs from those traditionally used in computer vision. Our CNN exploits both local features as well as more global contextual features simultaneously. Also, different from most traditional uses of CNNs, our networks use a final layer that is a convolutional implementation of a fully connected layer which allows a 40 fold speed up. We also describe a 2-phase training procedure that allows us to tackle difficulties related to the imbalance of tumor labels. Finally, we explore a cascade architecture in which the output of a basic CNN is treated as an additional source of information for a subsequent CNN. Results reported on the 2013 BRATS test data-set reveal that our architecture improves over the currently published state-of-the-art while being over 30 times faster. (C) 2016 Elsevier B.V. All rights reserved.