Treasure in Distribution: A Domain Randomization Based Multi-source Domain Generalization for 2D Medical Image Segmentation

doi:10.1007/978-3-031-43901-8_9

	Treasure in Distribution: A Domain Randomization Based Multi-source Domain Generalization for 2D Medical Image Segmentation
	Chen, Ziyang 1; Pan, Yongsheng2 ; Ye, Yiwen 1; Cui, Hengfei 1; Xia, Yong 1
	2023
会议录名称	LECTURE NOTES IN COMPUTER SCIENCE (INCLUDING SUBSERIES LECTURE NOTES IN ARTIFICIAL INTELLIGENCE AND LECTURE NOTES IN BIOINFORMATICS)
ISSN	0302-9743
卷号	14223 LNCS
页码	89-99
发表状态	已发表
DOI	10.1007/978-3-031-43901-8_9
摘要	Although recent years have witnessed the great success of convolutional neural networks (CNNs) in medical image segmentation, the domain shift issue caused by the highly variable image quality of medical images hinders the deployment of CNNs in real-world clinical applications. Domain generalization (DG) methods aim to address this issue by training a robust model on the source domain, which has a strong generalization ability. Previously, many DG methods based on feature-space domain randomization have been proposed, which, however, suffer from the limited and unordered search space of feature styles. In this paper, we propose a multi-source DG method called Treasure in Distribution (TriD), which constructs an unprecedented search space to obtain the model with strong robustness by randomly sampling from a uniform distribution. To learn the domain-invariant representations explicitly, we further devise a style-mixing strategy in our TriD, which mixes the feature styles by randomly mixing the augmented and original statistics along the channel wise and can be extended to other DG methods. Extensive experiments on two medical segmentation tasks with different modalities demonstrate that our TriD achieves superior generalization performance on unseen target-domain data. Code is available at https://github.com/Chen-Ziyang/TriD. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.
关键词	Convolutional neural networks Generative adversarial networks Image segmentation Medical imaging Mixing Random processes Convolutional neural network Deep learning Domain generalization Domain randomization Generalisation Medical image segmentation Multi-Sources Randomisation Real-world Search spaces
会议名称	26th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2023
出版地	GEWERBESTRASSE 11, CHAM, CH-6330, SWITZERLAND
会议地点	Vancouver, BC, Canada
会议日期	October 8, 2023 - October 12, 2023
URL	查看原文
收录类别	EI ; CPCI-S
语种	英语
资助项目	National Natural Science Foundation of China[62171377] ; Key Technologies Research and Development Program[2022YFC2009903/2022YFC2009900] ; Key Research and Development Program of Shaanxi Province, China[2022GY-084] ; China Postdoctoral Science Foundation[2021M703340/BX2021333]
WOS研究方向	Computer Science ; Radiology, Nuclear Medicine & Medical Imaging
WOS类目	Computer Science, Software Engineering ; Computer Science, Theory & Methods ; Radiology, Nuclear Medicine & Medical Imaging
WOS记录号	WOS:001109630700009
出版者	Springer Science and Business Media Deutschland GmbH
EI入藏号	20234314954835
EI主题词	Deep learning
EISSN	1611-3349
EI分类号	461.1 Biomedical Engineering ; 461.4 Ergonomics and Human Factors Engineering ; 723.4 Artificial Intelligence ; 746 Imaging Techniques ; 802.3 Chemical Operations ; 922.1 Probability Theory
原始文献类型	Conference article (CA)
引用统计	正在获取...
文献类型	会议论文
条目标识符	https://kms.shanghaitech.edu.cn/handle/2MSLDSTB/348726
专题	生物医学工程学院生物医学工程学院_PI研究组_沈定刚组
通讯作者	Xia, Yong
作者单位	1.Northwestern Polytech Univ, Sch Comp Sci & Engn, Natl Engn Lab Integrated Aerosp Ground Ocean Big, Xian 710072, Peoples R China 2.ShanghaiTech Univ, Sch Biomed Engn, Shanghai 201210, Peoples R China
推荐引用方式 GB/T 7714	Chen, Ziyang,Pan, Yongsheng,Ye, Yiwen,et al. Treasure in Distribution: A Domain Randomization Based Multi-source Domain Generalization for 2D Medical Image Segmentation[C]. GEWERBESTRASSE 11, CHAM, CH-6330, SWITZERLAND:Springer Science and Business Media Deutschland GmbH,2023:89-99.