IEEE Trans Med Imaging. 2020 Sep;39(9):2794-2805. doi: 10.1109/TMI.2020.2975853. Epub 2020 Feb 24.
Accurate segmentation of organs at risk (OARs) from head and neck (H&N) CT images is crucial for effective H&N cancer radiotherapy. However, the existing deep learning methods are often not trained in an end-to-end fashion, i.e., they independently predetermine the regions of target organs before organ segmentation, causing limited information sharing between related tasks and thus leading to suboptimal segmentation results. Furthermore, when conventional segmentation network is used to segment all the OARs simultaneously, the results often favor big OARs over small OARs. Thus, the existing methods often train a specific model for each OAR, ignoring the correlation between different segmentation tasks. To address these issues, we propose a new multi-view spatial aggregation framework for joint localization and segmentation of multiple OARs using H&N CT images. The core of our framework is a proposed region-of-interest (ROI)-based fine-grained representation convolutional neural network (CNN), which is used to generate multi-OAR probability maps from each 2D view (i.e., axial, coronal, and sagittal view) of CT images. Specifically, our ROI-based fine-grained representation CNN (1) unifies the OARs localization and segmentation tasks and trains them in an end-to-end fashion, and (2) improves the segmentation results of various-sized OARs via a novel ROI-based fine-grained representation. Our multi-view spatial aggregation framework then spatially aggregates and assembles the generated multi-view multi-OAR probability maps to segment all the OARs simultaneously. We evaluate our framework using two sets of H&N CT images and achieve competitive and highly robust segmentation performance for OARs of various sizes.
准确地从头颈部 (H&N) CT 图像中分割危及器官 (OAR) 对于有效的 H&N 癌症放射治疗至关重要。然而，现有的深度学习方法通常不是端到端训练的，即它们独立地预先确定目标器官的区域，然后再进行器官分割，从而导致相关任务之间的信息共享有限，从而导致分割结果不理想。此外，当使用传统的分割网络同时对所有 OAR 进行分割时，结果通常更有利于大 OAR 而不是小 OAR。因此，现有的方法通常为每个 OAR 训练一个特定的模型，而忽略了不同分割任务之间的相关性。为了解决这些问题，我们提出了一种新的多视图空间聚合框架，用于使用 H&N CT 图像联合定位和分割多个 OAR。我们框架的核心是一个基于感兴趣区域 (ROI) 的细粒度表示卷积神经网络 (CNN)，用于从 CT 图像的每个二维视图 (即轴向、冠状和矢状视图) 生成多 OAR 概率图。具体来说，我们的基于 ROI 的细粒度表示 CNN（1）统一了 OAR 定位和分割任务，并以端到端的方式对它们进行训练；（2）通过新颖的基于 ROI 的细粒度表示来改善各种大小的 OAR 的分割结果。然后，我们的多视图空间聚合框架对生成的多视图多 OAR 概率图进行空间聚合和组装，以同时分割所有 OAR。我们使用两组 H&N CT 图像评估了我们的框架，并获得了具有竞争力和高度稳健的各种大小 OAR 的分割性能。
