基于不确定性协作图和双层聚合机制的个性化联邦医学图像分割

杜晓刚; 魏征; 雷涛; 刘统飞; 王营博

doi:10.12263/DZXB.20250856

您当前的位置：

首页 >

文章列表页 >

基于不确定性协作图和双层聚合机制的个性化联邦医学图像分割

学术论文 | 更新时间：2026-06-16

- 基于不确定性协作图和双层聚合机制的个性化联邦医学图像分割
- Personalized Federated Medical Image Segmentation Based on Uncertain Collaboration Graph and Dual-Layer Aggregation Mechanism
- 电子学报 2026年54卷第3期页码：1078-1093
- 作者机构：
  
  1.陕西科技大学电子信息与人工智能学院，陕西西安 710021
  2.陕西科技大学陕西省人工智能联合实验室，陕西西安 710021
- 作者简介：
  
  杜晓刚男，1985年出生于陕西省宝鸡市。现为陕西科技大学电子信息与人工智能学院副教授。主要研究方向为机器学习、计算机视觉、医学图像处理。E-mail: duxiaogang@sust.edu.cn
  魏征男，2001年4月出生于河南省许昌市。现为陕西科技大学电子信息与人工智能学院硕士研究生。主要研究方向为联邦学习、医学图像分割。E-mail: 231612074@sust.edu.cn
  雷涛男，1981年11月出生于陕西省渭南市。2011年在西北工业大学获得博士学位，现为陕西科技大学电子信息与人工智能学院教授、博士生导师。主要研究方向为图像处理、模式识别和计算机视觉等。中国电子学会会员编号：E190184479M。E-mail: leitao@sust.edu.cn
  刘统飞男，1995年8月出生于陕西省榆林市。现为陕西科技大学电子信息与人工智能学院讲师、硕士生导师。主要研究方向为深度学习与医学、遥感图像智能解译。E-mail: liutongfei@sust.edu.cn
  王营博男，1990年3月出生于陕西省宝鸡市。2021年6月毕业于北京理工大学，现为陕西科技大学电子信息与人工智能学院讲师。主要研究方向为图像复原及增强。E-mail: wangyingbo@sust.edu.cn
- 基金信息：
  
  国家自然科学基金(62271296;62201334);西安市中青年科技创新领军人才项目(25ZQRC00019);陕西省创新能力支持计划(2025RS-CXTD-012);陕西省教育厅青年创新团队科研计划(23JP022;23JP014;25JP023)
- DOI：10.12263/DZXB.20250856
  中图分类号： TP391;
- 收稿：2025-09-28，
  
  录用：2026-02-25，
  
  纸质出版：2026-03-25
- 稿件说明：
移动端阅览
杜晓刚, 魏征, 雷涛, 等. 基于不确定性协作图和双层聚合机制的个性化联邦医学图像分割[J]. 电子学报, 2026, 54(03): 1078-1093.

DU Xiaogang, WEI Zheng, LEI Tao, et al. Personalized Federated Medical Image Segmentation Based on Uncertain Collaboration Graph and Dual-Layer Aggregation Mechanism[J]. Acta Electronica Sinica, 2026, 54(03): 1078-1093.
杜晓刚, 魏征, 雷涛, 等. 基于不确定性协作图和双层聚合机制的个性化联邦医学图像分割[J]. 电子学报, 2026, 54(03): 1078-1093. DOI：10.12263/DZXB.20250856

DU Xiaogang, WEI Zheng, LEI Tao, et al. Personalized Federated Medical Image Segmentation Based on Uncertain Collaboration Graph and Dual-Layer Aggregation Mechanism[J]. Acta Electronica Sinica, 2026, 54(03): 1078-1093. DOI：10.12263/DZXB.20250856

摘要

个性化联邦学习作为一种分布式机器学习范式，能够在不泄露客户端原始数据的前提下，实现多客户端模型的协同训练，已成为医学影像智能处理与分析领域的研究热点。然而，现有的个性化联邦学习方法主要通过全局协同或聚类分组协同来建模客户端关系，其整体协同粒度粗且灵活性不足。近年来，基于协作图的个性化联邦学习方法通过图结构建模客户端之间的协作关系，能够实现较细粒度的动态协同，有效缓解了全局协同与聚类协同的固有缺陷。但是，其仅以数据量和模型相似度来更新客户端协作图，未考虑医学图像分割任务中固有的高不确定性，导致其易受高不确定性客户端的影响而使得分割精度下降。为了解决该问题，提出了一种基于不确定性协作图和双层聚合机制的个性化联邦医学图像分割方法。该方法的核心优势主要包括两个方面：一是设计了不确定性惩罚项并将其引入服务器端目标函数中来优化协作图更新过程，生成适配医学图像分割任务的不确定性协作图，通过动态调整各个客户端之间的协作权重并避免高噪声参数混入导致知识污染，有效保障了协同训练的稳定性。二是提出了基于不确定性协作图的双层聚合机制。第一层聚合实现基于协作图的客户端局部协同，挖掘相似客户端之间的有效知识；第二层聚合通过融合局部协同结果与全局模型，来平衡全局模型通用性与本地客户端的个性化需求，实现了高质量知识的有效传递，提升了客户端本地模型的分割性能。为了全面验证所提方法的有效性与鲁棒性，在四个公开的息肉分割数据集上开展了大量的实验。实验结果表明：与其他先进的医学图像分割方法相比，提出的方法在多个客户端测试数据上取得了更优异的分割性能，为临床医疗场景下的个性化联邦医学图像分割提供了一种新的技术方案。

Abstract

As a distributed machine learning paradigm

personalized federated learning can realize the collaborative training of multi-client models without leaking the original data of the client

and has become a research hotspot in the field of medical image intelligent processing and analysis. However

the existing personalized federated learning methods mainly model client relationships through global collaboration or clustering group collaboration

and their overall collaboration granularity is coarse and lack of flexibility. In recent years

personalized federated learning methods based on collaboration graph model the collaboration relationship between clients by graph structure

which can achieve fine-grained dynamic collaboration and effectively alleviate the inherent defects of global collaboration and clustering collaboration. However

it only uses the amount of data and model similarity to update the client collaboration graph

and does not consider the inherent high uncertainty in the medical image segmentation task

which makes it vulnerable to high uncertainty clients and reduces the segmentation accuracy. In order to solve this problem

we propose a personalized federated medical image segmentation method based on uncertain collaboration graph and dual-layer aggregation mechanism in this paper. The core advantages of this method mainly include two aspects. Firstly

an uncertainty penalty term is designed and introduced into the server-side objective function to optimize the updating process of the collaboration graph

and generate an uncertain collaboration graph suitable for the medical image segmentation task. By dynamically adjusting the collaboration weights between each client and avoiding knowledge pollution caused by high noise parameters

the stability of collaborative training is effectively guaranteed. Secondly

a dual-layer aggregation mechanism based on uncertain collaboration graph is proposed. The first layer of aggregation realizes the local collaboration of clients based on collaboration graph

and mines the effective knowledge between similar clients. The second layer of aggregation balances the generality of the global model and the personalized requirements of the local client by fusing the local collaborative results and the global model

realizes the effective transfer of high-quality knowledge

and improves the segmentation performance of the client-side local model. In order to fully verify the effectiveness and robustness of the proposed method

a large number of experiments are carried out on four public polyp segmentation datasets. The experimental results show that compared with other advanced medical image segmentation methods

the proposed method achieves better segmentation performance on multiple client test data

which provides a new technical solution for personalized federal medical image segmentation in clinical medical scenarios.

关键词

Keywords

references

Ronneberger O , Fischer P , Brox T . U-Net: Convolutional networks for biomedical image segmentation [C ] // 18th International Conference on Medical Image Computing and Computer-Assisted Intervention . Heidelberg : Springer , 2015 : 234 - 241 . DOI: 10.1007/978-3-319-24574-4_28 http://dx.doi.org/10.1007/978-3-319-24574-4_28

Zhou Zongwei , Siddiquee M R , Tajbakhsh N , et al . UNet++: Redesigning skip connections to exploit multiscale features in image segmentation [J ] . IEEE Transactions on Medical Imaging , 2020 , 39 ( 6 ): 1856 - 1867 . DOI: 10.1109/TMI.2019.2959609 http://dx.doi.org/10.1109/TMI.2019.2959609

Chen J , Mei J , Li X , et al . TransUNet: Rethinking the U-Net architecture design for medical image segmentation through the lens of transformers [J ] . Medical Image Analysis , 2024 , 97 : 103280 . DOI: 10.1016/j.media.2024.103280 http://dx.doi.org/10.1016/j.media.2024.103280

Hatamizadeh A , Nath V , Tang Yucheng , et al . Swin UNETR: Swin transformers for semantic segmentation of brain tumors in MRI images [C ] // 7th International Workshop on Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries . Heidelberg : Springer , 2021 : 272 - 284 . DOI: 10.1007/978-3-031-08999-2_22 http://dx.doi.org/10.1007/978-3-031-08999-2_22

杨子瑶 , 雷涛 , 杜晓刚 , 等 . 基于可疑像素相互修正的半监督医学图像分割 [J ] . 电子学报 , 2025 , 53 ( 5 ): 1607 - 1621 .

Yang Ziyao , Lei Tao , Du Xiaogang , et al . Semi-supervised medical image segmentation based on suspicious pixel mutual correction [J ] . Acta Electronica Sinica , 2025 , 53 ( 5 ): 1607 - 1621 . (in Chinese)

雷涛 , 张峻铭 , 杜晓刚 , 等 . 基于混洗特征编码与门控解码的医学图像分割网络 [J ] . 电子学报 , 2024 , 52 ( 12 ): 4142 - 4152 . DOI: 10.12263/DZXB.20231011 http://dx.doi.org/10.12263/DZXB.20231011

Lei Tao , Zhang Junming , Du Xiaogang , et al . Medical image segmentation network based on shuffled feature encoding and gated decoding [J ] . Acta Electronica Sinica , 2024 , 52 ( 12 ): 4142 - 4152 . (in Chinese) . DOI: 10.12263/DZXB.20231011 http://dx.doi.org/10.12263/DZXB.20231011

McMahan B , Moore E , Ramage D , et al . Communication-efficient learning of deep networks from decentralized data [C ] // Proceedings of the 20th International Conference on Artificial Intelligence and Statistics . PMLR , 2017 : 1273 - 1282 .

Sheller M J , Edwards B , Reina G A , et al . Federated learning in medicine: Facilitating multi-institutional collaborations without sharing patient data [J ] . Scientific Reports , 2020 , 10 ( 1 ): 12598 . DOI: 10.1038/s41598-020-69250-1 http://dx.doi.org/10.1038/s41598-020-69250-1

Jiang Meirui , Roth H R , Li Wenqi , et al . Fair federated medical image segmentation via client contribution estimation [C ] // Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . Piscataway : IEEE , 2023 : 16302 - 16311 . DOI: 10.48550/arXiv.2303.16520 http://dx.doi.org/10.48550/arXiv.2303.16520

Ma Benteng , Zhang Jing , Xia Yong , et al . VNAS: Variational neural architecture search [J ] . International Journal of Computer Vision , 2024 , 132 ( 9 ): 3689 - 3713 . DOI: 10.1007/s11263-024-02014-w http://dx.doi.org/10.1007/s11263-024-02014-w

Zhang Ling , Wang Xiaosong , Yang Dong , et al . Generalizing deep learning for medical image segmentation to unseen domains via deep stacked transformation [J ] . IEEE Transactions on Medical Imaging , 2020 , 39 ( 7 ): 2531 - 2540 . DOI: 10.1109/TMI.2020.2973595 http://dx.doi.org/10.1109/TMI.2020.2973595

Fallah A , Mokhtari A , Ozdaglar A . Personalized federated learning with theoretical guarantees: A model-agnostic meta-learning approach [C ] // Proceedings of the 34th International Conference on Neural Information Processing Systems . New York : Curran Associates Inc. , 2020 : 300 .

Yu Tao , Bagdasaryan E , Shmatikov V . Salvaging federated learning by local adaptation [PP/OL ] . V3. arXiv ( 2022-03-03 )[ 2025-09-06 ] . https://arxiv.org/abs/2002.04758 https://arxiv.org/abs/2002.04758 .

Tan A Z , Yu Han , Cui Lizhen , et al . Towards personalized federated learning [J ] . IEEE Transactions on Neural Networks and Learning Systems , 2023 , 34 ( 12 ): 9587 - 9603 . DOI: 10.1109/tnnls.2022.3160699 http://dx.doi.org/10.1109/tnnls.2022.3160699

Makhija D , Ghosh J , Ho N . A Bayesian approach for personalized federated learning in heterogeneous settings [C ] // Proceedings of the 38th International Conference on Neural Information Processing Systems . New York : Curran Associates Inc. , 2024 : 3252 . DOI: 10.52202/079017-3252 http://dx.doi.org/10.52202/079017-3252

Deng Yuyang , Kamani M M , Mahdavi M . Adaptive personalized federated learning [C/OL ] // 9th International Conference on Learning Representations . 2021 : 1 - 47 . https://openreview.net/forum id=g0a-XYjpQ7r https://openreview.net/forumid=g0a-XYjpQ7r . DOI: 10.1109/JIOT.2023.3346900 http://dx.doi.org/10.1109/JIOT.2023.3346900

Arivazhagan M , Aggarwal V , Singh A K , et al . Federated Learning with Personalization Layers [PP/OL ] . V1.arXiv ( 2019-12-02 )[ 2025-09-06 ] . https://doi.org/10.48550/arXiv.1912.00818 https://doi.org/10.48550/arXiv.1912.00818 .

Ye Rui , Ni Zhenyang , Wu Fangzhao , et al . Personalized federated learning with inferred collaboration graphs [C ] // Proceedings of the 40th International Conference on Machine Learning . PMLR , 2023 : 39801 - 39817 . DOI: 10.1145/3529836.3529904 http://dx.doi.org/10.1145/3529836.3529904

Zhou Ziran , Gao Guanyu , Wu Xiaohu , et al . Personalized federated learning via learning dynamic graphs [PP/OL ] . V1.arXiv ( 2025-03-07 )[ 2025-09-06 ] . https://arxiv.org/abs/2503.05474 https://arxiv.org/abs/2503.05474 . DOI: 10.1016/b978-0-44-323641-9.00012-1 http://dx.doi.org/10.1016/b978-0-44-323641-9.00012-1

Liu Quande , Chen Cheng , Qin Jing , et al . FedDG: Federated domain generalization on medical image segmentation via episodic learning in continuous frequency space [C ] // Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2021 : 1013 - 1023 . DOI: 10.48550/arXiv.2103.06030 http://dx.doi.org/10.48550/arXiv.2103.06030

Li Xiaoxiao , Jiang Meirui , Zhang Xiaofei , et al . FedBN: Federated learning on non-IID features via local batch normalization [C/OL ] // 9th International Conference on Learning Representations . 2021 : 1 - 27 . https://openreview.net/forum id=6YEQUn0QICG https://openreview.net/forumid=6YEQUn0QICG . DOI: 10.48550/arXiv.2102.07623 http://dx.doi.org/10.48550/arXiv.2102.07623

Chen Jiayi , Ma Benteng , Cui Hengfei , et al . FedEvi: Improving federated medical image segmentation via evidential weight aggregation [C ] // 27th International Conference on Medical Image Computing and Computer-Assisted Intervention . Heidelberg : Springer , 2024 : 361 - 372 . DOI: 10.1007/978-3-031-72117-5_34 http://dx.doi.org/10.1007/978-3-031-72117-5_34

Siomos V , Passerat-Palmbach J , Tarroni G . FedCLAM: Client adaptive momentum with foreground intensity matching for federated medical image segmentation [C ] // 28th International Conference on Medical Image Computing and Computer Assisted Intervention . Heidelberg : Springer , 2025 : 247 - 257 . DOI: 10.1007/978-3-032-04978-0_24 http://dx.doi.org/10.1007/978-3-032-04978-0_24

Jiang Le , Ma Liyan , Zeng Tieyong , et al . UFPS: A unified framework for partially annotated federated segmentation in heterogeneous data distribution [J ] . Patterns , 2024 , 5 ( 2 ): 100917 . DOI: 10.1016/j.patter.2024.100917 http://dx.doi.org/10.1016/j.patter.2024.100917

Xu Xuanang , Deng H H , Gateno J , et al . Federated multi-organ segmentation with inconsistent labels [J ] . IEEE Transactions on Medical Imaging , 2023 , 42 ( 10 ): 2948 - 2960 . DOI: 10.1109/tmi.2023.3270140 http://dx.doi.org/10.1109/tmi.2023.3270140

Tölle M , Navarro F , Eble S , et al . FUNAvg: Federated uncertainty weighted averaging for datasets with diverse labels [C ] // 27th International Conference on Medical Image Computing and Computer Assisted Intervention . Heidelberg : Springer , 2024 : 405 - 415 . DOI: 10.1007/978-3-031-72117-5_38 http://dx.doi.org/10.1007/978-3-031-72117-5_38

Wu Nannan , Sun Zhaobin , Yan Zengqiang , et al . FedA 3 I: Annotation quality-aware aggregation for federated medical image segmentation against heterogeneous annotation noise[C ] // Proceedings of the AAAI Conference on Artificial Intelligence . 2024 , 38 ( 14 ): 15943 - 15951 . DOI: 10.1609/aaai.v38i14.29525 http://dx.doi.org/10.1609/aaai.v38i14.29525

Xiang Yangyang , Wu Nannan , Yu Li , et al . FedIA: Federated medical image segmentation with heterogeneous annotation completeness [C ] // 27th International Conference on Medical Image Computing and Computer Assisted Intervention . Heidelberg : Springer , 2024 : 373 - 382 . DOI: 10.1007/978-3-031-72117-5_35 http://dx.doi.org/10.1007/978-3-031-72117-5_35

Wang Jiacheng , Jin Yueming , Wang Liansheng . Personalizing federated medical image segmentation via local calibration [C ] // 17th European Conference on Computer Vision . Heidelberg : Springer , 2022 : 456 - 472 . DOI: 10.48550/arXiv.2207.04655 http://dx.doi.org/10.48550/arXiv.2207.04655

Wang Jiacheng , Jin Yueming , Stoyanov D , et al . FedDP: Dual personalization in federated medical image segmentation [J ] . IEEE Transactions on Medical Imaging , 2024 , 43 ( 1 ): 297 - 308 . DOI: 10.1109/TMI.2023.3299206 http://dx.doi.org/10.1109/TMI.2023.3299206

Jiang Meirui , Yang Hongzheng , Cheng Chen , et al . IOP-FL: Inside-outside personalization for federated medical image segmentation [J ] . IEEE Transactions on Medical Imaging , 2023 , 42 ( 7 ): 2106 - 2117 . DOI: 10.1109/tmi.2023.3263072 http://dx.doi.org/10.1109/tmi.2023.3263072

Xie Luyuan , Lin Manqing , Liu Siyuan , et al . pFLFE: Cross-silo personalized federated learning via feature enhancement on medical image segmentation [C ] // 27th International Conference on Medical Image Computing and Computer Assisted Intervention . Heidelberg : Springer , 2024 : 599 - 610 . DOI: 10.1007/978-3-031-72117-5_56 http://dx.doi.org/10.1007/978-3-031-72117-5_56

Sattler F , Müller K R , Samek W . Clustered federated learning: Model-agnostic distributed multitask optimization under privacy constraints [J ] . IEEE Transactions on Neural Networks and Learning Systems , 2021 , 32 ( 8 ): 3710 - 3722 . DOI: 10.1109/tnnls.2020.3015958 http://dx.doi.org/10.1109/tnnls.2020.3015958

Shamsian A , Navon A , Fetaya E , et al . Personalized federated learning using hypernetworks [C ] // Proceedings of the 38th International Conference on Machine Learning . PMLR , 2021 : 9489 - 9502 . DOI: 10.48550/arXiv.2103.04628 http://dx.doi.org/10.48550/arXiv.2103.04628

Bernal J , Sánchez J , Vilariño F . Towards automatic polyp detection with a polyp appearance model [J ] . Pattern Recognition , 2012 , 45 ( 9 ): 3166 - 3182 . DOI: 10.1016/j.patcog.2012.03.002 http://dx.doi.org/10.1016/j.patcog.2012.03.002

Bernal J , Sánchez F J , Fernández-Esparrach G , et al . WM-DOVA maps for accurate polyp highlighting in colonoscopy: Validation vs. saliency maps from physicians [J ] . Computerized Medical Imaging and Graphics , 2015 , 43 : 99 - 111 . DOI: 10.1016/j.compmedimag.2015.02.007 http://dx.doi.org/10.1016/j.compmedimag.2015.02.007

Silva J , Histace A , Romain O , et al . Toward embedded detection of polyps in WCE images for early diagnosis of colorectal cancer [J ] . International Journal of Computer Assisted Radiology and Surgery , 2014 , 9 ( 2 ): 283 - 293 . DOI: 10.1007/s11548-013-0926-3 http://dx.doi.org/10.1007/s11548-013-0926-3

Jha D , Smedsrud P H , Riegler M A , et al . Kvasir-SEG: A segmented polyp dataset [C ] // 26th International Conference on Multimedia Modeling . Heidelberg : Springer , 2020 : 451 - 462 . DOI: 10.1007/978-3-030-37734-2_37 http://dx.doi.org/10.1007/978-3-030-37734-2_37

Li T , Sahu A K , Zaheer M , et al . Federated optimization in heterogeneous networks [C ] // Proceedings of the Third Conference on Machine Learning and Systems . mlsys . org , 2020 : 429 - 450 . DOI: 10.1109/ieeeconf44664.2019.9049023 http://dx.doi.org/10.1109/ieeeconf44664.2019.9049023

Collins L , Hassani H , Mokhtari A , et al . Exploiting shared representations for personalized federated learning [C ] // Proceedings of the 38th International Conference on Machine Learning . PMLR , 2021 : 2089 - 2099 . DOI: 10.48550/arXiv.2102.07078 http://dx.doi.org/10.48550/arXiv.2102.07078

Oh J , Kim S , Yun S Y . FedBABU: Toward enhanced representation for federated image classification [C/OL ] // 10th International Conference on Learning Representations . 2022 : 1 - 29 . https://openreview.net/forum id=HuaYQfggn5u https://openreview.net/forumid=HuaYQfggn5u .

Yao Dezhong , Pan Wanning , Dai Yutong , et al . FedGKD: Toward heterogeneous federated learning via global knowledge distillation [J ] . IEEE Transactions on Computers , 2024 , 73 ( 1 ): 3 - 17 . DOI: 10.1109/TC.2023.3315066 http://dx.doi.org/10.1109/TC.2023.3315066

Li Tian , Hu Shengyuan , Beirami A , et al . Ditto: Fair and robust federated learning through personalization [C ] // Proceedings of the 38th International Conference on Machine Learning . PMLR , 2021 : 6357 - 6368 .

Lin Li , Liu Yixiang , Wu Jiewei , et al . FedLPPA: Learning personalized prompt and aggregation for federated weakly-supervised medical image segmentation [J ] . IEEE Transactions on Medical Imaging , 2025 , 44 ( 3 ): 1127 - 1139 . DOI: 10.1109/tmi.2024.3483221 http://dx.doi.org/10.1109/tmi.2024.3483221

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于可疑像素相互修正的半监督医学图像分割

面向不确定性融合的多模态遥感数据分类

基于交叉视觉状态空间与多分支交互注意力的医学图像分割

MSGPose：基于多语义图卷积与图引导状态空间模型的单目单人三维人体姿态估计

非平衡有向图下欧拉-拉格朗日群体智能系统分布式时变优化