Frequency‐aware domain randomization for single‐source domain generalization in medical image segmentation
The proposed RandDG method effectively addresses the critical challenges of texture bias and limited style diversity in single-source DG for medical image segmentation through frequency-aware dual-space randomization framework. This method shows promise for practical deployment in clinical settings, where target domain data may not always be available.
Authors
- Jingmin Xin (ORCID: https://orcid.org/0000-0003-1906-2327)
- Ziteng Cui (ORCID: https://orcid.org/0000-0002-6126-2712)
- Shaoyi Du
- Jiayi Wu (ORCID: https://orcid.org/0000-0001-5371-8989)
- Zikai Chen (ORCID: https://orcid.org/0000-0003-4113-100X)
- Yiyi Chen (ORCID: https://orcid.org/0000-0002-1228-5803)
- Xiaofeng Yang (ORCID: https://orcid.org/0000-0001-9920-4641)
Institutions
- Emory University (US)
- China Aerospace Science and Technology Corporation (CN)
- Xi'an Jiaotong University (CN)
Publication Details
- Journal
- Medical Physics
- Published
- 2025-11-29
- DOI
- https://doi.org/10.1002/mp.70118
- Primary Topic
- Domain Adaptation and Few-Shot Learning
- Type
- article
- Field-Weighted Citation Impact
- 0.00