Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
97 tokens/sec
GPT-4o
53 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

COSMOS: Cross-Modality Unsupervised Domain Adaptation for 3D Medical Image Segmentation based on Target-aware Domain Translation and Iterative Self-Training (2203.16557v2)

Published 30 Mar 2022 in eess.IV and cs.CV

Abstract: Recent advances in deep learning-based medical image segmentation studies achieve nearly human-level performance when in fully supervised condition. However, acquiring pixel-level expert annotations is extremely expensive and laborious in medical imaging fields. Unsupervised domain adaptation can alleviate this problem, which makes it possible to use annotated data in one imaging modality to train a network that can successfully perform segmentation on target imaging modality with no labels. In this work, we propose a self-training based unsupervised domain adaptation framework for 3D medical image segmentation named COSMOS and validate it with automatic segmentation of Vestibular Schwannoma (VS) and cochlea on high-resolution T2 Magnetic Resonance Images (MRI). Our target-aware contrast conversion network translates source domain annotated T1 MRI to pseudo T2 MRI to enable segmentation training on target domain, while preserving important anatomical features of interest in the converted images. Iterative self-training is followed to incorporate unlabeled data to training and incrementally improve the quality of pseudo-labels, thereby leading to improved performance of segmentation. COSMOS won the 1\textsuperscript{st} place in the Cross-Modality Domain Adaptation (crossMoDA) challenge held in conjunction with the 24th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2021). It achieves mean Dice score and Average Symmetric Surface Distance of 0.871(0.063) and 0.437(0.270) for VS, and 0.842(0.020) and 0.152(0.030) for cochlea.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (32)
  1. Synergistic image and feature adaptation: Towards cross-modality domain adaptation for medical image segmentation. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 33, pages 865–872, 2019.
  2. Jae Won Choi. Using out-of-the-box frameworks for unpaired image translation and image segmentation for the crossmoda challenge. arXiv e-prints, pages arXiv–2110, 2021.
  3. Distribution matching losses can hallucinate features in medical image translation. In International conference on medical image computing and computer-assisted intervention, pages 529–536. Springer, 2018.
  4. Unsupervised domain adaptation in semantic segmentation based on pixel alignment and self-training. arXiv preprint arXiv:2109.14219, 2021.
  5. Scribble-based domain adaptation via co-segmentation. In International Conference on Medical Image Computing and Computer-Assisted Intervention, pages 479–489. Springer, 2020.
  6. Crossmoda 2021 challenge: Benchmark of cross-modality domain adaptation techniques for vestibular schwnannoma and cochlea segmentation. arXiv preprint arXiv:2201.02831, 2022.
  7. Unsupervised domain adaptation by backpropagation. In International conference on machine learning, pages 1180–1189. PMLR, 2015.
  8. Cycada: Cycle-consistent adversarial domain adaptation. In International conference on machine learning, pages 1989–1998. PMLR, 2018.
  9. Synseg-net: Synthetic segmentation without target modality ground truth. IEEE transactions on medical imaging, 38(4):1016–1025, 2018.
  10. nnu-net: a self-configuring method for deep learning-based biomedical image segmentation. Nature methods, 18(2):203–211, 2021.
  11. Psigan: joint probabilistic segmentation and image distribution matching for unpaired cross-modality adaptation-based mri segmentation. IEEE Transactions on Medical Imaging, 39(12):4071–4084, 2020.
  12. Self-training for end-to-end speech recognition. In ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 7084–7088. IEEE, 2020.
  13. Learning texture invariant representation for domain adaptation of semantic segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 12975–12984, 2020.
  14. Bidirectional learning for domain adaptation of semantic segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 6936–6945, 2019.
  15. Fang Liu. Susan: segment unannotated image structure using adversarial network. Magnetic resonance in medicine, 81(5):3330–3345, 2019.
  16. Cross-modality domain adaptation for vestibular schwannoma and cochlea segmentation. arXiv preprint arXiv:2109.06274, 2021.
  17. Cycle self-training for domain adaptation. arXiv preprint arXiv:2103.03571, 2021.
  18. A closer look at self-training for zero-label semantic segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 2693–2702, 2021.
  19. Segmentation of vestibular schwannoma from mri, an open annotated dataset and baseline algorithm. Scientific Data, 8(1):1–6, 2021.
  20. Segmentation of vestibular schwannoma from mri—an open annotated dataset and baseline algorithm. medRxiv, 2021.
  21. An artificial intelligence framework for automatic segmentation and volumetry of vestibular schwannomas from contrast-enhanced t1-weighted and high-resolution t2-weighted mri. Journal of neurosurgery, 134(1):171–179, 2019.
  22. Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results. arXiv preprint arXiv:1703.01780, 2017.
  23. Learning to adapt structured output space for semantic segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 7472–7481, 2018.
  24. Adversarial discriminative domain adaptation. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 7167–7176, 2017.
  25. Self-training with noisy student improves imagenet classification. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 10687–10698, 2020.
  26. Billion-scale semi-supervised learning for image classification. arXiv preprint arXiv:1905.00546, 2019.
  27. Translating and segmenting multimodal medical volumes with cycle-and shape-consistency generative adversarial network. In Proceedings of the IEEE conference on computer vision and pattern Recognition, pages 9242–9251, 2018.
  28. Unpaired image-to-image translation using cycle-consistent adversarial networks. In Proceedings of the IEEE international conference on computer vision, pages 2223–2232, 2017.
  29. Improving semantic segmentation via self-training. arXiv preprint arXiv:2004.14960, 2020.
  30. Rethinking pre-training and self-training. arXiv preprint arXiv:2006.06882, 2020.
  31. Unsupervised domain adaptation for semantic segmentation via class-balanced self-training. In Proceedings of the European conference on computer vision (ECCV), pages 289–305, 2018.
  32. Pseudoseg: Designing pseudo labels for semantic segmentation. arXiv preprint arXiv:2010.09713, 2020.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (6)
  1. Hyungseob Shin (5 papers)
  2. Hyeongyu Kim (3 papers)
  3. Sewon Kim (5 papers)
  4. Yohan Jun (9 papers)
  5. Taejoon Eo (3 papers)
  6. Dosik Hwang (6 papers)
Citations (23)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now