Automatic delineation of organ-at-risk (OAR) and gross-tumor-volume (GTV) is of great significance for radiotherapy planning. However, it is a challenging task to learn powerful representations for accurate delineation under limited pixel (voxel)-wise annotations. Contrastive learning at pixel-level can alleviate the dependency on annotations by learning dense representations from unlabeled data. Recent studies in this direction design various contrastive losses on the feature maps, to yield discriminative features for each pixel in the map. However, pixels in the same map inevitably share semantics to be closer than they actually are, which may affect the discrimination of pixels in the same map and lead to the unfair comparison to pixels in other maps. To address these issues, we propose a separated region-level contrastive learning scheme, namely SepaReg, the core of which is to separate each image into regions and encode each region separately. Specifically, SepaReg comprises two components: a structure-aware image separation (SIS) module and an intra- and inter-organ distillation (IID) module. The SIS is proposed to operate on the image set to rebuild a region set under the guidance of structural information. The inter-organ representation will be learned from this set via typical contrastive losses cross regions. On the other hand, the IID is proposed to tackle the quantity imbalance in the region set as tiny organs may produce fewer regions, by exploiting intra-organ representations. We conducted extensive experiments to evaluate the proposed model on a public dataset and two private datasets. The experimental results demonstrate the effectiveness of the proposed model, consistently achieving better performance than state-of-the-art approaches. Code is available at https://github.com/jcwang123/Separate_CL.
翻译:自动划定器官风险(OAR)和毛图卷(GTV)对于放射治疗规划非常重要,然而,在有限的像素(voxel)的注释下,为准确划界而学习强大的表达方式是一项具有挑战性的任务。像素层面的对比学习可以通过从未贴标签的数据中学习密集的表达方式减轻对说明的依赖。这个方向的最近研究设计了地貌图上的各种对比性损失,为地图中的每个像素产生有区别的特征。然而,同一地图中的像素不可避免地会分享比实际更接近的语义,这可能会影响同一地图中的像素的差别,导致与其他地图中的像素的不公示意图进行不公的比较。为了解决这些问题,我们提议在区域一级进行不同的对比学习方案,即SepaReg,其核心是将每个图像分离到各个区域,并分别对每个区域进行编码。具体地说,SepaReg由两个组成部分组成:结构-了解图像分离(SIS)模块和内部和组织间对等方的表达(II-D)图像的表达方式,在Silentalalalalal 模块下,将显示一个Salideal-deal dealalalalal deal deal deal demodustral deal deal deal demodudududududududududududustr rodude the sal rodudududustr ro disal) ro ro ro ro ro ro ro ro dism dismal dism disal rodudududududududududududududude the rod roduction rod rout ro ro rod the ro ro rod ro ro ro rodudududududeal ro rod rod rod ro ro rod rod rodal rodal rodal rodal roal rodal rod ro roal rod rod rod rod rod rodal rodal rodal rodal roal roal rodal ro ro ro ro ro ro