Scaling Tumor Segmentation: Best Lessons from Real and Synthetic Data
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912682755489792 |
|---|---|
| author | Chen, Qi Zhou, Xinze Liu, Chen Chen, Hao Li, Wenxuan Jiang, Zekun Huang, Ziyan Zhao, Yuxuan Yu, Dexin He, Junjun Zheng, Yefeng Shao, Ling Yuille, Alan Zhou, Zongwei |
| author_facet | Chen, Qi Zhou, Xinze Liu, Chen Chen, Hao Li, Wenxuan Jiang, Zekun Huang, Ziyan Zhao, Yuxuan Yu, Dexin He, Junjun Zheng, Yefeng Shao, Ling Yuille, Alan Zhou, Zongwei |
| contents | AI for tumor segmentation is limited by the lack of large, voxel-wise annotated datasets, which are hard to create and require medical experts. In our proprietary JHH dataset of 3,000 annotated pancreatic tumor scans, we found that AI performance stopped improving after 1,500 scans. With synthetic data, we reached the same performance using only 500 real scans. This finding suggests that synthetic data can steepen data scaling laws, enabling more efficient model training than real data alone. Motivated by these lessons, we created AbdomenAtlas 2.0--a dataset of 10,135 CT scans with a total of 15,130 tumor instances per-voxel manually annotated in six organs (pancreas, liver, kidney, colon, esophagus, and uterus) and 5,893 control scans. Annotated by 23 expert radiologists, it is several orders of magnitude larger than existing public tumor datasets. While we continue expanding the dataset, the current version of AbdomenAtlas 2.0 already provides a strong foundation--based on lessons from the JHH dataset--for training AI to segment tumors in six organs. It achieves notable improvements over public datasets, with a +7% DSC gain on in-distribution tests and +16% on out-of-distribution tests. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_14831 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Scaling Tumor Segmentation: Best Lessons from Real and Synthetic Data Chen, Qi Zhou, Xinze Liu, Chen Chen, Hao Li, Wenxuan Jiang, Zekun Huang, Ziyan Zhao, Yuxuan Yu, Dexin He, Junjun Zheng, Yefeng Shao, Ling Yuille, Alan Zhou, Zongwei Computer Vision and Pattern Recognition AI for tumor segmentation is limited by the lack of large, voxel-wise annotated datasets, which are hard to create and require medical experts. In our proprietary JHH dataset of 3,000 annotated pancreatic tumor scans, we found that AI performance stopped improving after 1,500 scans. With synthetic data, we reached the same performance using only 500 real scans. This finding suggests that synthetic data can steepen data scaling laws, enabling more efficient model training than real data alone. Motivated by these lessons, we created AbdomenAtlas 2.0--a dataset of 10,135 CT scans with a total of 15,130 tumor instances per-voxel manually annotated in six organs (pancreas, liver, kidney, colon, esophagus, and uterus) and 5,893 control scans. Annotated by 23 expert radiologists, it is several orders of magnitude larger than existing public tumor datasets. While we continue expanding the dataset, the current version of AbdomenAtlas 2.0 already provides a strong foundation--based on lessons from the JHH dataset--for training AI to segment tumors in six organs. It achieves notable improvements over public datasets, with a +7% DSC gain on in-distribution tests and +16% on out-of-distribution tests. |
| title | Scaling Tumor Segmentation: Best Lessons from Real and Synthetic Data |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2510.14831 |