Zhu Hui, Shu Shi, Zhang Jianping
School of Mathematics and Computational Science, Xiangtan University, Xiangtan, 411105, China.
School of Computational Science and Electronics, Hunan Institute of Engineering, Xiangtan, 411104, China.
Med Biol Eng Comput. 2025 Feb;63(2):429-446. doi: 10.1007/s11517-024-03208-7. Epub 2024 Oct 4.
Segmentation of organs at risks (OARs) in the thorax plays a critical role in radiation therapy for lung and esophageal cancer. Although automatic segmentation of OARs has been extensively studied, it remains challenging due to the varying sizes and shapes of organs, as well as the low contrast between the target and background. This paper proposes a cascaded FAS-UNet+ framework, which integrates convolutional neural networks and nonlinear multi-grid theory to solve a modified Mumford-shah model for segmenting OARs. This framework is equipped with an enhanced iteration block, a coarse-to-fine multiscale architecture, an iterative optimization strategy, and a model ensemble technique. The enhanced iteration block aims to extract multiscale features, while the cascade module is used to refine coarse segmentation predictions. The iterative optimization strategy improves the network parameters to avoid unfavorable local minima. An efficient data augmentation method is also developed to train the network, which significantly improves its performance. During the prediction stage, a weighted ensemble technique combines predictions from multiple models to refine the final segmentation. The proposed cascaded FAS-UNet+ framework was evaluated on the SegTHOR dataset, and the results demonstrate significant improvements in Dice score and Hausdorff Distance (HD). The Dice scores were 95.22%, 95.68%, and HD values were 0.1024, and 0.1194 for the segmentations of the aorta and heart in the official unlabeled dataset, respectively. Our code and trained models are available at https://github.com/zhuhui100/C-FASUNet-plus .
胸部危及器官（OARs）的分割在肺癌和食管癌的放射治疗中起着关键作用。尽管已经对OARs的自动分割进行了广泛研究，但由于器官大小和形状各异，以及目标与背景之间的对比度较低，分割仍然具有挑战性。本文提出了一种级联FAS-UNet+框架，该框架集成了卷积神经网络和非线性多网格理论，以求解用于分割OARs的修正Mumford-shah模型。该框架配备了增强迭代块、从粗到细的多尺度架构、迭代优化策略和模型集成技术。增强迭代块旨在提取多尺度特征，而级联模块用于细化粗分割预测。迭代优化策略改进网络参数以避免不利的局部最小值。还开发了一种有效的数据增强方法来训练网络，这显著提高了其性能。在预测阶段，加权集成技术结合多个模型的预测来细化最终分割。所提出的级联FAS-UNet+框架在SegTHOR数据集上进行了评估，结果表明在Dice分数和豪斯多夫距离（HD）方面有显著改进。在官方未标记数据集中，主动脉和心脏分割的Dice分数分别为95.22%、95.68%，HD值分别为0.1024和0.1194。我们的代码和训练模型可在https://github.com/zhuhui100/C-FASUNet-plus获取。
