Abstract Deep neural networks have been widely used for medical image analysis. However, the lack of access to a large-scale annotated dataset poses a great challenge, especially in the case of rare diseases or new domains for the research society. Transfer of pre-trained features from the relatively large dataset is a considerable solution. In this paper, we have explored supervised segmentation using domain adaptation for optic nerve and orbital tumour, when only small sampled CT images are given. Even the lung image database consortium image collection (LIDC-IDRI) is a cross-domain to orbital CT, but the proposed domain adaptation method improved the performance of attention U-Net for the segmentation in public optic nerve dataset and our clinical orbital tumour dataset by 3.7% and 13.7% in the Dice score, respectively. The code and dataset are available at https://github.com/cmcbigdata.

中文翻译：

---

小样本轨道CT图像的域自适应监督分割

摘要 深度神经网络已广泛用于医学图像分析。然而，无法访问大规模注释数据集带来了巨大挑战，特别是在罕见疾病或研究社会的新领域的情况下。从相对较大的数据集中转移预训练的特征是一个相当不错的解决方案。在本文中，我们探索了使用域自适应对视神经和眼眶肿瘤进行监督分割，当仅给出小样本 CT 图像时。即使是肺部图像数据库联盟图像收集（LIDC-IDRI）也是一个跨域到眼眶 CT，但所提出的域适应方法提高了注意力 U-Net 在公共视神经数据集和我们的临床眼眶肿瘤数据集中分割的性能骰子得分分别提高了 3.7% 和 13.7%。