Deep Multi-Class Segmentation Without Ground-Truth Labels
Deep Multi-Class Segmentation Without Ground-Truth Labels
复制标题
DOI：
--
复制
发表时间：
2018-04
期刊：
影响因子：
--
通讯作者：
T. Joyce;A. Chartsias;S. Tsaftaris
中科院分区：
文献类型：
--
作者：
T. Joyce;A. Chartsias;S. Tsaftaris
在本文中，我们证明，通过使用对抗性训练和额外的无监督成本，可以训练多类解剖分割算法，而无需为要分割的数据集提供任何真实标签。具体来说，使用来自相同解剖结构的不同数据集的标签（尽管可能采用不同的模态），我们训练一个模型，通过对抗性学习，从 CT 和 MRI 中的输入心脏图像合成真实的多通道标签掩模。然而，正如预期的那样，生成真实的掩模图像本身不足以完成分割任务：模型可以使用输入图像作为噪声源并合成高度真实的分割掩模，这些掩模不一定在空间上与输入相对应。为了克服这个问题，我们引入了额外的无监督成本，并证明这些成本提供了足够的进一步指导来产生良好的分割结果。我们在多模态全心脏分割挑战 (MM-WHS) [1] 的 CT 和 MR 数据上测试了我们提出的方法，并展示了与没有监督成本的变体相比，我们的无监督成本对改善分割结果的影响。
In this paper we demonstrate that through the use of adversarial training and additional unsupervised costs it is possible to train a multi-class anatomical segmentation algorithm without any ground-truth labels for the data set to be segmented. Specifically, using labels from a different data set of the same anatomy (although potentially in a different modality) we train a model to synthesise realistic multi-channel label masks from input cardiac images in both CT and MRI, through adversarial learning. However, as is to be expected, generating realistic mask images is not, on its own, sufficient for the segmentation task: the model can use the input image as a source of noise and synthesise highly realistic segmentation masks that do no necessarily correspond spatially to the input. To overcome this, we introduce additional unsupervised costs, and demonstrate that these provide sufficient further guidance to produce good segmentation results. We test our proposed method on both CT and MR data from the multi-modal whole heart segmentation challenge (MM-WHS) [1], and show the effect of our unsupervised costs on improving the segmentation results, in comparison to a variant without them.