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

Primitive Geometry Segment Pre-training for 3D Medical Image Segmentation (2401.03665v1)

Published 8 Jan 2024 in cs.CV

Abstract: The construction of 3D medical image datasets presents several issues, including requiring significant financial costs in data collection and specialized expertise for annotation, as well as strict privacy concerns for patient confidentiality compared to natural image datasets. Therefore, it has become a pressing issue in 3D medical image segmentation to enable data-efficient learning with limited 3D medical data and supervision. A promising approach is pre-training, but improving its performance in 3D medical image segmentation is difficult due to the small size of existing 3D medical image datasets. We thus present the Primitive Geometry Segment Pre-training (PrimGeoSeg) method to enable the learning of 3D semantic features by pre-training segmentation tasks using only primitive geometric objects for 3D medical image segmentation. PrimGeoSeg performs more accurate and efficient 3D medical image segmentation without manual data collection and annotation. Further, experimental results show that PrimGeoSeg on SwinUNETR improves performance over learning from scratch on BTCV, MSD (Task06), and BraTS datasets by 3.7%, 4.4%, and 0.3%, respectively. Remarkably, the performance was equal to or better than state-of-the-art self-supervised learning despite the equal number of pre-training data. From experimental results, we conclude that effective pre-training can be achieved by looking at primitive geometric objects only. Code and dataset are available at https://github.com/SUPER-TADORY/PrimGeoSeg.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (32)
  1. The medical segmentation decathlon. Nature communications, 13(1):4128, 2022.
  2. Billion-scale pretraining with vision transformers for multi-task visual representations. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pages 564–573, 2022.
  3. Med3d: Transfer learning for 3d medical image analysis. arXiv preprint arXiv:1904.00625, 2019.
  4. Improved baselines with momentum contrastive learning. arXiv preprint arXiv:2003.04297, 2020.
  5. Masked image modeling advances 3d medical image analysis. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pages 1970–1980, 2023.
  6. Niftynet: a deep-learning platform for medical imaging. Computer methods and programs in biomedicine, 158:113–122, 2018.
  7. Self-supervised pretraining of visual features in the wild. arXiv preprint arXiv:2103.01988, 2021.
  8. Transferable visual words: Exploiting the semantics of anatomical patterns for self-supervised learning. IEEE transactions on medical imaging, 40(10):2857–2868, 2021.
  9. Swin unetr: Swin transformers for semantic segmentation of brain tumors in mri images. In Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries: 7th International Workshop, BrainLes 2021, Held in Conjunction with MICCAI 2021, Virtual Event, September 27, 2021, Revised Selected Papers, Part I, pages 272–284. Springer, 2022a.
  10. Unetr: Transformers for 3d medical image segmentation. In Proceedings of the IEEE/CVF winter conference on applications of computer vision, pages 574–584, 2022b.
  11. Initialization using perlin noise for training networks with a limited amount of data. In 2020 25th International Conference on Pattern Recognition (ICPR), pages 1023–1028. IEEE, 2021.
  12. Self-supervised 3d anatomy segmentation using self-distilled masked image transformer (smit). In International Conference on Medical Image Computing and Computer-Assisted Intervention, pages 556–566, Cham, September 2022. Springer Nature Switzerland.
  13. Pre-training without natural images. In Proceedings of the Asian Conference on Computer Vision, 2020.
  14. Formula-driven supervised learning with recursive tiling patterns. In Proceedings of the IEEE/CVF International Conference on Computer Vision Workshops, pages 4098–4105, October 2021.
  15. Replacing labeled real-image datasets with auto-generated contours. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 21232–21241, 2022a.
  16. Spatiotemporal initialization for 3d cnns with generated motion patterns. 2022b.
  17. Big transfer (bit): General visual representation learning. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part V 16, pages 491–507. Springer, 2020.
  18. Miccai multi-atlas labeling beyond the cranial vault– workshop and challenge. 2015.
  19. Decoupled weight decay regularization. arXiv preprint arXiv:1711.05101, 2017.
  20. Can vision transformers learn without natural images? arXiv preprint arXiv:2103.13023, 2021.
  21. Joint self-supervised image-volume representation learning with intra-inter contrastive clustering. arXiv preprint arXiv:2212.01893, 2022.
  22. Pre-training auto-generated volumetric shapes for 3d medical image segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, pages 4739–4744, June 2023.
  23. Visual atoms: Pre-training vision transformers with sinusoidal waves, 2023.
  24. 3d self-supervised methods for medical imaging. Advances in neural information processing systems, 33:18158–18172, 2020.
  25. Self-supervised pre-training of swin transformers for 3d medical image analysis. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 20730–20740, 2022.
  26. Unimiss: Universal medical self-supervised learning via breaking dimensionality barrier. In European Conference on Computer Vision, pages 558–575, Cham, October 2022. Springer Nature Switzerland.
  27. Point cloud pre-training with natural 3d structures. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 21283–21293, 2022.
  28. Desd: Self-supervised learning with deep self-distillation for 3d medical image segmentation. In Medical Image Computing and Computer Assisted Intervention–MICCAI 2022: 25th International Conference, Singapore, September 18–22, 2022, Proceedings, Part IV, pages 545–555. Springer, 2022.
  29. Momentum contrastive voxel-wise representation learning for semi-supervised volumetric medical image segmentation. In Medical Image Computing and Computer Assisted Intervention–MICCAI 2022: 25th International Conference, Singapore, September 18–22, 2022, Proceedings, Part IV, pages 639–652. Springer, 2022.
  30. Preservational learning improves self-supervised medical image models by reconstructing diverse contexts. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 3499–3509, 2021a.
  31. Models genesis. Medical image analysis, 67:101840, 2021b.
  32. Rubik’s cube+: A self-supervised feature learning framework for 3d medical image analysis. Medical image analysis, 64:101746, 2020.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (5)
  1. Ryu Tadokoro (4 papers)
  2. Ryosuke Yamada (10 papers)
  3. Kodai Nakashima (5 papers)
  4. Ryo Nakamura (11 papers)
  5. Hirokatsu Kataoka (55 papers)
Citations (3)
View on Semantic Scholar

Summary

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

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

GitHub

  1. GitHub - SUPER-TADORY/PrimGeoSeg (6 stars)