Glasgow College, University of Electronic Science and Technology of China, Chengdu, 611731, China.
Department of Radiology, Peking University People's Hospital, Beijing, 100044, China.
Med Phys. 2021 Nov;48(11):6987-7002. doi: 10.1002/mp.15260. Epub 2021 Oct 18.
Radiotherapy is one of the main treatments of nasopharyngeal cancer (NPC) and lung cancer. Accurate segmentation of organs at risks (OARs) in CT images is a key step in radiotherapy planning for NPC and lung cancer. However, the segmentation of OARs is influenced by the highly imbalanced size of organs, which often results in very poor segmentation results for small and difficult-to-segment organs. In addition, the complex morphological changes and fuzzy boundaries of OARs also pose great challenges to the segmentation task. In this paper, we propose a cross-layer attention fusion network (CLAF-CNN) to solve the problem of accurately segmenting OARs.
In CLAF-CNN, we integrate the spatial attention maps of the adjacent spatial attention modules to make the segmentation targets more accurately focused, so that the network can capture more target-related features. In this way, the spatial attention modules in the network can be learned and optimized together. In addition, we introduce a new Top-K exponential logarithmic Dice loss (TELD-Loss) to solve the imbalance problem in OAR segmentation. The TELD-Loss further introduces a Top-K optimization mechanism based on Dice loss and exponential logarithmic loss, which makes the network pay more attention to small organs and difficult-to-segment organs, so as to enhance the overall performance of the segmentation model.
We validated our framework on the OAR segmentation datasets of the head and neck and lung CT images in the StructSeg 2019 challenge. Experiments show that the CLAF-CNN outperforms the state-of-the-art attention-based segmentation methods in the OAR segmentation task with average Dice coefficient of 79.65% for head and neck OARs and 88.39% for lung OARs.
This work provides a new network named CLAF-CNN which contains cross-layer spatial attention map fusion architecture and TELD-Loss for OAR segmentation. Results demonstrated that the proposed method could obtain accurate segmentation results for OARs, which has a potential of improving the efficiency of radiotherapy planning for nasopharynx cancer and lung cancer.
放射治疗是鼻咽癌（NPC）和肺癌的主要治疗方法之一。在 NPC 和肺癌的放射治疗计划中，准确地对危及器官（OARs）进行分割是关键步骤。然而，OAR 的分割受到器官高度不平衡大小的影响，这通常导致对小且难以分割的器官的分割结果非常差。此外，OAR 的复杂形态变化和模糊边界也给分割任务带来了巨大的挑战。在本文中，我们提出了一种跨层注意融合网络（CLAF-CNN）来解决准确分割 OAR 的问题。
在 CLAF-CNN 中，我们整合相邻空间注意模块的空间注意图，使分割目标更准确地聚焦，从而使网络能够捕获更多与目标相关的特征。通过这种方式，网络中的空间注意模块可以一起学习和优化。此外，我们引入了一种新的 Top-K 指数对数 Dice 损失（TELD-Loss）来解决 OAR 分割中的不平衡问题。TELD-Loss 进一步引入了基于 Dice 损失和指数对数损失的 Top-K 优化机制，使网络更加关注小器官和难以分割的器官，从而增强分割模型的整体性能。
我们在 2019 年 StructSeg 挑战赛的头颈部和肺部 CT 图像 OAR 分割数据集上验证了我们的框架。实验表明，CLAF-CNN 在 OAR 分割任务中优于基于注意力的最新分割方法，对头颈部 OAR 的平均 Dice 系数为 79.65%，对肺部 OAR 的平均 Dice 系数为 88.39%。
这项工作提供了一种名为 CLAF-CNN 的新网络，该网络包含跨层空间注意图融合架构和 TELD-Loss 用于 OAR 分割。结果表明，所提出的方法可以获得 OAR 的准确分割结果，这有可能提高鼻咽癌和肺癌的放射治疗计划的效率。
