Collaborative Learning of Augmentation and Disentanglement for Semi-Supervised Domain Generalized Medical Image Segmentation
- 6 August 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 45 (1) , 281-295
Abstract
This paper explores a challenging yet realistic scenario: semi-supervised domain generalization (SSDG) that includes label scarcity and domain shift problems. We pinpoint that the limitations of previous SSDG methods lie in 1) neglecting the difference between domain shifts existing within a training dataset (intra-domain shift, IDS) and those occurring between training and testing datasets (cross-domain shift, CDS) and 2) overlooking the interplay between label scarcity and domain shifts, resulting in these methods merely stitching together semi-supervised learning (SSL) and domain generalization (DG) techniques. Considering these limitations, we propose a novel perspective to decompose SSDG into the combination of unsupervised domain adaptation (UDA) and DG problems. To this end, we design a causal augmentation and disentanglement framework (CausalAD) for semi-supervised domain generalized medical image segmentation. Concretely, CausalAD involves two collaborative processes: an augmentation process, which utilizes disentangled style factors to perform style augmentation for UDA, and a disentanglement process, which decouples domain-invariant (content) and domain-variant (noise and style) features for DG. Furthermore, we propose a proxy-based self-paced training strategy (ProSPT) to guide the training of CausalAD by gradually selecting unlabeled image pixels with high-quality pseudo labels in a self-paced training manner. Finally, we introduce a hierarchical structural causal model (HSCM) to explain the intuition and concept behind our method. Extensive experiments in the cross-sequence, cross-site, and cross-modality semi-supervised domain generalized medical image segmentation settings show the effectiveness of CausalAD and its superiority over the state-of-the-art. The code is available at https://github.com/Senyh/CausalAD.Keywords
Funding Information
- National Natural Science Foundation of China (62076059)
- National Natural Science Foundation of China (U22A2022)
- Science and Technology Joint Project of Liaoning province (2023JH2/101700367)
- Fundamental Research Funds for the Central Universities (N25BJD005)
- China Scholarship Council (202406080040)
This publication has 44 references indexed in Scilit:
- Semi-Supervised Domain Generalization with Stochastic StyleMatchInternational Journal of Computer Vision, 2023
- Disentangle domain features for cross-modality cardiac image segmentationMedical Image Analysis, 2021
- Image-Level Harmonization of Multi-site Data Using Image-and-Spatial Transformer Networks2020
- Kvasir-SEG: A Segmented Polyp Dataset2019
- Evaluation of algorithms for Multi-Modality Whole Heart Segmentation: An open-access grand challengeMedical Image Analysis, 2019
- A theoretical understanding of self-paced learningInformation Sciences, 2017
- Empirical Risk Minimization2017
- Multi-scale patch and multi-modality atlases for whole heart segmentation of MRIMedical Image Analysis, 2016
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015
- WM-DOVA maps for accurate polyp highlighting in colonoscopy: Validation vs. saliency maps from physiciansComputerized Medical Imaging and Graphics, 2015