---
title: "Continual Unsupervised Domain Adaptation for Cardiac Image Segmentation with Style-Adapting Generative Replay and Prototype Consolidation"
authors:
  - "Shun Xiang"
  - "Yan Yi"
  - "Haiyong Chen"
  - "Yuanquan Wang"
  - "Yining Wang"
doi: "10.24963/ijcai.2026/210"
url: "https://ijcai.org/proceedings/2026/210"
pdf: "https://ijcai.org/proceedings/2026/0210.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "1884-1892"
track: "Main Track"
keywords:
  - "Computer Vision: Transfer, low-shot, semi- and un- supervised learning"
  - "Computer Vision: Biomedical image analysis"
  - "Computer Vision: Segmentation, grouping and shape analysis"
---

# Continual Unsupervised Domain Adaptation for Cardiac Image Segmentation with Style-Adapting Generative Replay and Prototype Consolidation

Shun Xiang, Yan Yi, Haiyong Chen, Yuanquan Wang, Yining Wang

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 1884-1892. DOI: [10.24963/ijcai.2026/210](https://doi.org/10.24963/ijcai.2026/210)

## Abstract

Continual unsupervised domain adaptation (UDA) for multi-domain cardiac image segmentation is crucial for clinical deployment under strict privacy constraints, where data from different centers cannot be stored or revisited. However, existing methods still suffer from catastrophic forgetting and training instability caused by noisy pseudo-labels and representation drift. In this study, we propose a novel continual UDA framework that enforces dual-level alignment at the input and feature levels. At the input level, the Domain-Style Adapting Generative Replay (DSA-GR) module employs trajectory-consistent diffusion distillation and style adaptation to synthesize high-fidelity, domain-consistent replay images without storing raw data. At the feature level, the Cross-domain Prototype-guided Feature Consolidation (CPFC) mechanism mitigates representation drift by maintaining a cross-domain prototype memory bank that aggregates features from the current and replay streams. These prototypes act as robust anchors for prototype-guided contrastive learning, rectifying the feature space and suppressing pseudo-label drift. Extensive experiments on two public datasets demonstrate that our framework consistently outperforms state-of-the-art methods. Code will be released at https://github.com/hlyf-xs/CUDA_Seg.