Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2507.13087 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918137156337664 |
|---|---|
| author | Zhang, Han Luo, Xiangde Chen, Yong Li, Kang |
| author_facet | Zhang, Han Luo, Xiangde Chen, Yong Li, Kang |
| contents | Annotation variability remains a substantial challenge in medical image segmentation, stemming from ambiguous imaging boundaries and diverse clinical expertise. Traditional deep learning methods producing single deterministic segmentation predictions often fail to capture these annotator biases. Although recent studies have explored multi-rater segmentation, existing methods typically focus on a single perspective -- either generating a probabilistic ``gold standard'' consensus or preserving expert-specific preferences -- thus struggling to provide a more omni view. In this study, we propose DiffOSeg, a two-stage diffusion-based framework, which aims to simultaneously achieve both consensus-driven (combining all experts' opinions) and preference-driven (reflecting experts' individual assessments) segmentation. Stage I establishes population consensus through a probabilistic consensus strategy, while Stage II captures expert-specific preference via adaptive prompts. Demonstrated on two public datasets (LIDC-IDRI and NPC-170), our model outperforms existing state-of-the-art methods across all evaluated metrics. Source code is available at https://github.com/string-ellipses/DiffOSeg . |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_13087 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DiffOSeg: Omni Medical Image Segmentation via Multi-Expert Collaboration Diffusion Model Zhang, Han Luo, Xiangde Chen, Yong Li, Kang Computer Vision and Pattern Recognition Annotation variability remains a substantial challenge in medical image segmentation, stemming from ambiguous imaging boundaries and diverse clinical expertise. Traditional deep learning methods producing single deterministic segmentation predictions often fail to capture these annotator biases. Although recent studies have explored multi-rater segmentation, existing methods typically focus on a single perspective -- either generating a probabilistic ``gold standard'' consensus or preserving expert-specific preferences -- thus struggling to provide a more omni view. In this study, we propose DiffOSeg, a two-stage diffusion-based framework, which aims to simultaneously achieve both consensus-driven (combining all experts' opinions) and preference-driven (reflecting experts' individual assessments) segmentation. Stage I establishes population consensus through a probabilistic consensus strategy, while Stage II captures expert-specific preference via adaptive prompts. Demonstrated on two public datasets (LIDC-IDRI and NPC-170), our model outperforms existing state-of-the-art methods across all evaluated metrics. Source code is available at https://github.com/string-ellipses/DiffOSeg . |
| title | DiffOSeg: Omni Medical Image Segmentation via Multi-Expert Collaboration Diffusion Model |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2507.13087 |