Achieving fair medical image segmentation in foundation models with adversarial visual prompt tuning

doi:10.1016/j.ins.2025.122501

	Achieving fair medical image segmentation in foundation models with adversarial visual prompt tuning
	Li, Yuqi 1; Li, Yanli 2; Zhang, Kai 3; Zhang, Fuyan ; Yang, Chuanguang ; Guo, Zhongliang 4; Ding, Weiping 2,5; Huang, Tingwen 3
	2025-12-01
发表期刊	INFORMATION SCIENCES
ISSN	0020-0255
卷号	720 页码:16
摘要	Recent advances in deep learning have significantly enhanced medical image analysis capabilities. Medical image segmentation, a critical application in this domain, enables precise delineation of anatomical structures and pathological regions, substantially supporting clinical decision-making. However, current segmentation methods primarily optimize for overall performance without considering disparities across demographic groups, raising important fairness concerns. To address this gap, we propose Adversarial Visual Prompt Tuning (AdvVPT), a parameter-efficient approach that enhances fairness in foundation models for medical image segmentation. AdvVPT introduces trainable visual prompts within the image encoder while keeping the backbone frozen, requiring only 0.812M additional parameters. These prompts are optimized through adversarial training to absorb demographic-specific biased information from image embeddings, achieved by maximizing prediction errors for sensitive attributes and increasing embedding distances between visual prompts and image features. Experimental evaluation on the Harvard-FairSeg dataset demonstrates that AdvVPT achieves state-of-the-art fairness performance across multiple demographic attributes. For racial fairness, AdvVPT achieves an ES-Dice score of 0.8996 and an ES-IoU score of 0.8222 on optic cup segmentation, substantially outperforming existing methods. For gender fairness using the SAT backbone, AdvVPT achieves an ES-Dice of 0.9297 and ES-IoU of 0.8614, demonstrating both superior performance and improved balance between male and female subgroups.
关键词	Fairness Image segmentation Foundation models Visual prompt tuning Medicine data analysis
DOI	10.1016/j.ins.2025.122501
收录类别	SCI
语种	英语
资助项目	National Key RD Plan of China[2024YFE0202700]
WOS研究方向	Computer Science
WOS类目	Computer Science, Information Systems
WOS记录号	WOS:001533716700004
出版者	ELSEVIER SCIENCE INC
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/42035
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Ding, Weiping; Huang, Tingwen
作者单位	1.Chinese Acad Sci, Inst Comp Technol, Beijing, Peoples R China 2.Nantong Univ, Sch Artificial Intelligence & Comp Sci, Nantong 226019, Peoples R China 3.Shenzhen Univ Adv Technol, Shenzhen 518107, Peoples R China 4.Univ St Andrews, Sch Comp Sci, St Andrews, Scotland 5.City Univ Macau, Fac Data Sci, Macau 999078, Peoples R China
推荐引用方式 GB/T 7714	Li, Yuqi,Li, Yanli,Zhang, Kai,et al. Achieving fair medical image segmentation in foundation models with adversarial visual prompt tuning[J]. INFORMATION SCIENCES,2025,720:16.
APA	Li, Yuqi.,Li, Yanli.,Zhang, Kai.,Zhang, Fuyan.,Yang, Chuanguang.,...&Huang, Tingwen.(2025).Achieving fair medical image segmentation in foundation models with adversarial visual prompt tuning.INFORMATION SCIENCES,720,16.
MLA	Li, Yuqi,et al."Achieving fair medical image segmentation in foundation models with adversarial visual prompt tuning".INFORMATION SCIENCES 720(2025):16.