IS2Net: Intra-domain Semantic and Inter-domain Style Enhancement for Semi-supervised Medical Domain Generalization
Open Access
- 26 October 2023
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 8285-8293
Abstract
Domain generalization (DG) demonstrates superior generalization ability in cross-center medical image segmentation. Despite its great success, existing fully supervised DG methods require collecting a large quantity of pixel-level annotations which is quite expensive and time-consuming. To address this challenge, several semi-supervised domain generalized (SSDG) methods have been proposed by simply coupling semi-supervised learning (SSL) with DG tasks, which give rise to two main concerns: (1) Intra-domain dubious semantic information: the quality of pseudo labels in each source domain suffers from the limited amount of labeled data and cross-domain discrepancy. (2) Inter-domain intangible style relationship: current models fail in integrating domain-level information and overlook the relationships among different domains, which degrades the generalization ability of model. In light of these two issues, we propose a novel SSDG framework, namely IS2Net, by arranging an inter-domain generalization branch and several intra-domain SSL branches in a parallel manner, powered by two appealing designs that build a positive interaction between them: (1) A style and semantic memory mechanism is designed to provide both high-quality class-wise representations for intra-domain semantic enhancement and stable domain-specific knowledge for inter-domain style relationship construction. (2) Confident pseudo labeling strategy aims at generating more reliable supervision for intra and inter domain branches, and thus facilitating the learning process of the whole framework. Extensive experiments show that IS2Net yields consistent improvements over the state-of- the-art methods in three public benchmarks.Keywords
Funding Information
- Major Technological Innovation Project of Hangzhou (2022AIZD0147)
- National Key Research and Development Project (2022YFC2504605)
- Major Scientific Research Project of Zhejiang Lab (2020ND8AD01)
- Zhejiang Provincial Natural Science Foundation of China (LZ22F020012)
This publication has 19 references indexed in Scilit:
- Domain-Adversarial Learning for Multi-Centre, Multi-Vendor, and Multi-Disease Cardiac MR Image Segmentation2021
- Disentangled Representations for Domain-Generalized Cardiac Segmentation2021
- Shape-Aware Meta-learning for Generalizing Prostate MRI Segmentation to Unseen Domains2020
- Learning to Learn Single Domain Generalization2020
- UNet++: Redesigning Skip Connections to Exploit Multiscale Features in Image SegmentationIEEE Transactions on Medical Imaging, 2019
- CutMix: Regularization Strategy to Train Strong Classifiers With Localizable Features2019
- Domain Generalization by Solving Jigsaw Puzzles2019
- Domain-Specific Batch Normalization for Unsupervised Domain Adaptation2019
- Sigmoid-weighted linear units for neural network function approximation in reinforcement learningNeural Networks, 2018
- Domain Generalization with Adversarial Feature Learning2018