Pang Shuchao, Du Anan, Orgun Mehmet A, Yu Zhenmei, Wang Yunyun, Wang Yan, Liu Guanfeng
Department of Computing, Macquarie University, Sydney, NSW, 2109, Australia.
School of Electrical and Data Engineering, University of Technology Sydney, Ultimo, NSW, 2007, Australia.
Eur J Nucl Med Mol Imaging. 2020 Sep;47(10):2248-2268. doi: 10.1007/s00259-020-04781-3. Epub 2020 Mar 28.
Unlike the normal organ segmentation task, automatic tumor segmentation is a more challenging task because of the existence of similar visual characteristics between tumors and their surroundings, especially on computed tomography (CT) images with severe low contrast resolution, as well as the diversity and individual characteristics of data acquisition procedures and devices. Consequently, most of the recently proposed methods have become increasingly difficult to be applied on a different tumor dataset with good results, and moreover, some tumor segmentors usually fail to generalize beyond those datasets and modalities used in their original evaluation experiments.
In order to alleviate some of the problems with the recently proposed methods, we propose a novel unified and end-to-end adversarial learning framework for automatic segmentation of any kinds of tumors from CT scans, called CTumorGAN, consisting of a Generator network and a Discriminator network. Specifically, the Generator attempts to generate segmentation results that are close to their corresponding golden standards, while the Discriminator aims to distinguish between generated samples and real tumor ground truths. More importantly, we deliberately design different modules to take into account the well-known obstacles, e.g., severe class imbalance, small tumor localization, and the label noise problem with poor expert annotation quality, and then use these modules to guide the CTumorGAN training process by utilizing multi-level supervision more effectively.
We conduct a comprehensive evaluation on diverse loss functions for tumor segmentation and find that mean square error is more suitable for the CT tumor segmentation task. Furthermore, extensive experiments with multiple evaluation criteria on three well-established datasets, including lung tumor, kidney tumor, and liver tumor databases, also demonstrate that our CTumorGAN achieves stable and competitive performance compared with the state-of-the-art approaches for CT tumor segmentation.
In order to overcome those key challenges arising from CT datasets and solve some of the main problems existing in the current deep learning-based methods, we propose a novel unified CTumorGAN framework, which can be effectively generalized to address any kinds of tumor datasets with superior performance.
与正常器官分割任务不同，自动肿瘤分割是一项更具挑战性的任务，这是因为肿瘤与其周围组织存在相似的视觉特征，尤其是在对比度分辨率极低的计算机断层扫描（CT）图像上，以及数据采集程序和设备的多样性和个体特征。因此，最近提出的大多数方法越来越难以在不同的肿瘤数据集上取得良好效果，而且，一些肿瘤分割器通常无法推广到其原始评估实验中使用的数据集和模态之外。
为了缓解最近提出的方法存在的一些问题，我们提出了一种新颖的统一端到端对抗学习框架，用于从CT扫描中自动分割任何类型的肿瘤，称为CTumorGAN，它由一个生成器网络和一个判别器网络组成。具体而言，生成器试图生成接近其相应金标准的分割结果，而判别器旨在区分生成的样本和真实肿瘤的地面真值。更重要的是，我们特意设计了不同的模块来考虑众所周知的障碍，例如严重的类别不平衡、小肿瘤定位以及专家注释质量差导致的标签噪声问题，然后通过更有效地利用多级监督，使用这些模块来指导CTumorGAN的训练过程。
我们对用于肿瘤分割的各种损失函数进行了全面评估，发现均方误差更适合CT肿瘤分割任务。此外，在包括肺肿瘤、肾肿瘤和肝肿瘤数据库在内的三个成熟数据集上进行的具有多个评估标准的广泛实验也表明，与用于CT肿瘤分割的最新方法相比，我们的CTumorGAN实现了稳定且具有竞争力的性能。
为了克服CT数据集带来的那些关键挑战并解决当前基于深度学习的方法中存在的一些主要问题，我们提出了一种新颖的统一CTumorGAN框架，该框架可以有效地推广以处理任何类型的肿瘤数据集，并具有卓越的性能。
