Structure-Driven Unsupervised Domain Adaptation for Cross-Modality Cardiac Segmentation

Zhiming Cui, Changjian Li, Zhixu Du, Nenglun Chen, Guodong Wei, Runnan Chen, Lei Yang, Dinggang Shen, Wenping Wang

Abstract

Performance degradation due to domain shift remains a major challenge in medical image analysis. Unsupervised domain adaptation that transfers knowledge learned from the source domain with ground truth labels to the target domain without any annotation is the mainstream solution to resolve this issue. In this paper, we present a novel unsupervised domain adaptation framework for cross-modality cardiac segmentation, by explicitly capturing a common cardiac structure embedded across different modalities to guide cardiac segmentation. In particular, we first extract a set of 3D landmarks, in a self-supervised manner, to represent the cardiac structure of different modalities. The high-level structure information is then combined with another complementary feature, the Canny edges, to produce accurate cardiac segmentation results both in the source and target domains. We extensively evaluate our method on the MICCAI 2017 MM-WHS dataset for cardiac segmentation. The evaluation, comparison and comprehensive ablation studies demonstrate that our approach achieves satisfactory segmentation results and outperforms state-of-the-art unsupervised domain adaptation methods by a significant margin.
BibTeX
@article{cui2021structure,
  author = {Zhiming Cui and Changjian Li and Zhixu Du and Nenglun Chen and Guodong Wei and Runnan Chen and Lei Yang and Dinggang Shen and Wenping Wang},
  title = {{Structure-Driven Unsupervised Domain Adaptation for Cross-Modality Cardiac Segmentation}},
  journal = {IEEE Trans. Medical Imaging},
  year = {2021},
  url = {https://openreview.net/forum?id=cktvDkcs0xt},
}

Loading related content...

Figure