Deep learning of the sectional appearances of 3D CT images for anatomical structure segmentation based on an FCN voting method
Deep learning of the sectional appearances of 3D CT images for anatomical structure segmentation based on an FCN voting method
复制标题
DOI：
10.1002/mp.12480
复制
发表时间：
2017-10-01
期刊：
影响因子：
3.8
通讯作者：
Fujita, Hiroshi
中科院分区：
文献类型：
--
作者：
Zhou, Xiangrong;Takayama, Ryosuke;Fujita, Hiroshi
目的：我们提出了一个通过像素到标签深度学习训练的单一网络，以解决三维（3D）计算机断层扫描（CT）图像中自动多器官分割的一般问题。我们的方法可以描述为一种体素多类分类方案，用于自动为 2D/3D CT 图像中的每个像素/体素分配标签。 方法：我们将 CT 图像（通常为 3D）中解剖结构（包括多个器官）的分割算法简化为对从不同视点绘制的多个 2D 切片进行冗余语义分割的多数投票方案。该方法继承了由卷积层和反卷积层组成的用于 2D 语义图像分割的全卷积网络（FCN）的精神，并通过 3D-2D-3D 变换扩展了核心结构以适应 3D CT 图像分割。所提出的网络中的所有参数都是从少量 CT 病例中进行像素到标签训练，并以人类注释为基础事实。所提出的网络自然地满足了不同尺寸的CT病例中多器官分割的要求，无需任何调整即可覆盖任意扫描区域。结果：使用人体躯干的19个解剖结构的同时分割来训练和验证所提出的网络，包括17个主要器官和两个特殊区域（管腔和胃内内容物）。其中一些结构在之前的 CT 分割研究中从未报道过。我们的实验中使用了由 240 个（95% 用于训练，5% 用于测试）3D CT 扫描及其手动注释的地面实况分割组成的数据库。结果表明，相对于地面实况，19 个感兴趣的结构以可接受的精度进行了分割（训练和测试数据集中的体素分别为 88.1% 和 87.9% 被正确标记）。结论：我们提出了一种基于像素到标签深度学习的单一网络，以解决 3D CT 病例中解剖结构分割的挑战性问题。这项工作的新颖之处在于对 CT 病例的 3D 解剖结构的不同 2D 截面外观进行深度学习的策略，以及来自多个交叉 2D 截面的 3D 分割结果的多数投票，以实现可用性和可靠性，比必须由人类专业知识指导的传统分割方法具有更高的效率、通用性和灵活性。 (C) 2017 年作者。 《医学物理学》由 Wiley periodicals, Inc. 代表美国医学物理学家协会出版。
Purpose: We propose a single network trained by pixel-to-label deep learning to address the general issue of automatic multiple organ segmentation in three-dimensional (3D) computed tomography (CT) images. Our method can be described as a voxel-wise multiple-class classification scheme for automatically assigning labels to each pixel/voxel in a 2D/3D CT image.Methods: We simplify the segmentation algorithms of anatomical structures (including multiple organs) in a CT image (generally in 3D) to a majority voting scheme over the semantic segmentation of multiple 2D slices drawn from different viewpoints with redundancy. The proposed method inherits the spirit of fully convolutional networks (FCNs) that consist of convolution and deconvolution layers for 2D semantic image segmentation, and expands the core structure with 3D-2D-3D transformations to adapt to 3D CT image segmentation. All parameters in the proposed network are trained pixel-to-label from a small number of CT cases with human annotations as the ground truth. The proposed network naturally fulfills the requirements of multiple organ segmentations in CT cases of different sizes that cover arbitrary scan regions without any adjustment.Results: The proposed network was trained and validated using the simultaneous segmentation of 19 anatomical structures in the human torso, including 17 major organs and two special regions (lumen and content inside of stomach). Some of these structures have never been reported in previous research on CT segmentation. A database consisting of 240 (95% for training and 5% for testing) 3D CT scans, together with their manually annotated ground-truth segmentations, was used in our experiments. The results show that the 19 structures of interest were segmented with acceptable accuracy (88.1% and 87.9% voxels in the training and testing datasets, respectively, were labeled correctly) against the ground truth.Conclusions: We propose a single network based on pixel-to-label deep learning to address the challenging issue of anatomical structure segmentation in 3D CT cases. The novelty of this work is the policy of deep learning of the different 2D sectional appearances of 3D anatomical structures for CT cases and the majority voting of the 3D segmentation results from multiple crossed 2D sections to achieve availability and reliability with better efficiency, generality, and flexibility than conventional segmentation methods, which must be guided by human expertise. (C) 2017 The Authors. Medical Physics published by Wiley Periodicals, Inc. on behalf of American Association of Physicists in Medicine.