{"type":"ScholarlyArticle","title":"CoDG-Net: Structure-Guided Style Diffusion and Collaborative Learning to Mitigate Catastrophic Forgetting in Medical Image Domain Generalization","authors":[{"name":"Yucheng Song","slug":"yucheng-song","given":"Yucheng","family":"Song","affiliation":""},{"name":"Jincan Wang","slug":"jincan-wang","given":"Jincan","family":"Wang","affiliation":""},{"name":"Haokang Ding","slug":"haokang-ding","given":"Haokang","family":"Ding","affiliation":""},{"name":"Zhiqiang Tian","slug":"zhiqiang-tian","given":"Zhiqiang","family":"Tian","affiliation":""},{"name":"Kangxu Fan","slug":"kangxu-fan","given":"Kangxu","family":"Fan","affiliation":""},{"name":"Zhifang Liao","slug":"zhifang-liao","given":"Zhifang","family":"Liao","affiliation":""}],"abstract":"Domain Generalization (DG) for medical image segmentation is both highly challenging and critically important. However, existing medical DG methods largely overlook the issue of Catastrophic Forgetting (CF): Models often sacrifice their ability to retain source-domain knowledge while pursuing cross-domain robustness. This can directly threaten diagnostic safety in already-deployed clinical scenarios. To address this, we investigate data augmentation strategies and catastrophic forgetting for medical image DG segmentation. First, we propose a structure-guided style diffusion augmentation method. Constrained by anatomical structure consistency in the frequency domain, this method performs cross-domain diffusion on the amplitude spectrum, generating samples with more diverse and broader style coverage to better support domain generalization. Then, we design a collaborative learning network with a dual-branch interactive architecture (CoDG-Net), together with a novel learning bias-guided strategy that adaptively regulates knowledge transfer at both the layer level and the task level, thereby effectively mitigating catastrophic forgetting on the source domain. Experiments and ablation studies on single-source and multi-source medical DG benchmark datasets demonstrate that CoDG-Net not only outperforms existing state-of-the-art methods in target-domain segmentation performance, but also achieves a lower forgetting rate on the source-domain data. The code is available at: https://github.com/wangprocess/CoDG-Net.","doi":"10.24963/ijcai.2026/181","url":"https://ijcai.org/proceedings/2026/181","pdf":"https://ijcai.org/proceedings/2026/0181.pdf","markdown":"https://ijcai.org/proceedings/2026/181.md","bibtex":"https://ijcai.org/proceedings/2026/bibtex/181","volume":{"slug":"2026","title":"Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2026","url":"https://ijcai.org/proceedings/2026/","year":2026,"location":"Bremen, Germany","dates":"15-21 August 2026","editors":"Diego Calvanese","isbn":"978-1-956792-09-6","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2026"},"section":"Computer Vision","track":"Main Track","number":181,"pages":{"first":1622,"last":1630},"keywords":["Computer Vision: Segmentation, grouping and shape analysis","Computer Vision: Biomedical image analysis","Computer Vision: Transfer, low-shot, semi- and un- supervised learning"],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}