Subnetwork-based federated few-shot semantic segmentation of organ images

Abstract

Federated learning (FL), as a distributed learning paradigm, allows multiple medical institutions to collaborate on learning without the need to centralize all client data. However, existing methods pay little attention to more challenging medical image semantic segmentation tasks, especially in the scenario of the imbalanced dataset in federated few-shot learning (FSL). In this paper, we propose a subnetwork-based federated few-shot organ image segmentation method. Firstly, individual clients train using local training samples and then upload local model gradients to the server. The server utilizes their respective local model gradients to update the subnetwork maintained on the server and generate aggregation weights for forming personalized model parameters. Through this method, we can learn the similarities between different clients to address data heterogeneity issues. In addition, to enhance the communication efficiency between clients and the server, we have also designed a personalized layer aggregation strategy, which only transmits partial layer model parameters during the communication process to improve communication efficiency. Finally, we conducted experiments on abdomen magnetic resonance imaging (ABD-MRI) and abdomen computed tomography (ABD-CT) datasets to demonstrate the effectiveness of our method.

Keywords

A

References

[1] Bai H, Wang X, Guan Y, et al.. Blood cell counting based on U-Net++ and YOLOv5. Optoelectronics letters, 2023, 19(6): 370-376 J]

[2] Jiang J, Guo L, Liu W, et al.. Automatic diagnosis of multiple fundus lesions based on depth graph neural network. Optoelectronics letters, 2023, 19(5): 307-315 J]

[3] MCMAHAN B, MOORE E, RAMAGE D, et al. Communication-efficient learning of deep networks from decentralized data[EB/OL]. (2016-02-17) [2025-07-28]. https://arxiv.org/abs/1602.05629v3.

[4] LIU Z, WU F, WANG Y, et al. FedCL: federated contrastive learning for multi-center medical image classification[J]. Pattern recognition, 2023: 109739.

[5] Wang J, Jin Y, Stoyanov D, et al.. FedDP: dual personalization in federated medical image segmentation. IEEE transactions on medical imaging, 2024, 1: 43[J]

[6] Xiang Y, Wu N, Yu L, et al.. FedIA: federated medical image segmentation with heterogeneous annotation completeness. Medical Image Computing and Computer Assisted Intervention, October 6–10, 2024, Marrakesh, Morocco, 2024ChamSpringer373-382[C]

[7] Qin T, Fan C, Huang Y, et al.. pFLFE: cross-silo personalized federated learning via feature enhancement on medical image segmentation. Medical Image Computing and Computer Assisted Intervention, October 6–10, 2024, Marrakesh, Morocco, 2024ChamSpringer596-605[C]

[8] SHENG T, SHEN C, LIU Y, et al. Modeling global distribution for federated learning with label distribution skew[J]. Pattern recognition, 2023: 109724.

[9] Kim S, Park H, Chikontwe P, et al.. Communication efficient federated learning for multi-organ segmentation via knowledge distillation with image synthesis. IEEE transactions on medical imaging, 2025, 44(5): 2079-2092 J]

[10] Li T, Sahu A K, Zaheer M, et al.. Federated optimization in heterogeneous networks. Proceedings of machine learning and systems, 2020, 2: 429-450[J]

[11] ZHANG Y, GAO Y, DUAN H, et al. FedSCA: federated tuning with similarity-guided collaborative aggregation for heterogeneous medical image segmentation[EB/OL]. (2025-03-19) [2025-07-28]. https://arxiv.org/abs/2503.15390.

[12] Zhang H, Wang P, Liu J, et al.. pFedBCC: personalizing federated multi-target domain adaptive segmentation via bi-pole collaborative calibration. Computer methods and programs in biomedicine, 2025, 264: 108635 J]

[13] LI X, JIANG M, ZHANG X, et al. FedBN: federated learning on non-IID features via local batch normalization[EB/OL]. (2021-02-15) [2025-07-28]. https://arxiv.org/abs/2102.07623.

[14] Chen X, Yuan X. Decentralized personalization in federated medical image segmentation via gossip contrastive mutual learning. IEEE transactions on medical imaging, 2025, 44(4): 1659-1674[J]

[15] LU W, WANG J, CHEN Y, et al. Personalized federated learning with adaptive BatchNorm for healthcare[J]. IEEE transactions on big data, 2022.

[16] Tan Y, Long G, Liu L, et al.. FedProto: federated prototype learning across heterogeneous clients. Proceedings of the AAAI Conference on Artificial Intelligence, February 22–March 1, 2022, Virtual, 20228432-8440[C]

[17] Wang K, Liew J H, Zou Y, et al.. PANet: few-shot image semantic segmentation with prototype alignment. Proceedings of the IEEE/CVF International Conference on Computer Vision, October 27–November 2, 2019, Seoul, South Korea, 2019New YorkIEEE9197-9206[C]

[18] LIN Y, CHEN Y, CHENG K T, et al. Few shot medical image segmentation with cross attention transformer[EB/OL]. (2023-03-24) [2025-07-28]. https://arxiv.org/abs/2303.13867.

[19] Ding H, Sun C, Tang H, et al.. Few-shot medical image segmentation with cycle-resemblance attention. Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, January 2–7, 2023, Waikoloa, USA, 2023New YorkIEEE2488-2497[C]

[20] SIAM M, ORESHKIN B, JAGERSAND M. Adaptive masked proxies for few-shot segmentation[EB/OL]. (2019-02-19) [2025-07-28]. https://arxiv.org/abs/1902.11123?context=cs.

[21] Ouyang C, Biffi C, Chen C, et al.. Self-supervision with superpixels: training few-shot medical image segmentation without annotation. 16th European Conference on Computer Vision, August 23–28, 2020, Glasgow, UK, 2020ChamSpringer International Publishing762-780[C]

[22] Yu Q, Dang K, Tajbakhsh N, et al.. A location-sensitive local prototype network for few-shot medical image segmentation. IEEE 18th International Symposium on Biomedical Imaging, April 13–16, 2021, Virtual, 2021New YorkIEEE262-266[C]

[23] Fan C, Huang J. Federated few-shot learning with adversarial learning. 19th International Symposium on Modeling and Optimization in Mobile, Ad hoc, and Wireless Networks, October 18–21, 2021, Philadelphia, PA, USA, 2021New YorkIEEE1-8[C]

[24] LANDMAN B, XU Z, IGELSIAS J, et al. MICCAI multi-atlas labeling beyond the cranial vault-workshop and challenge (2015)[EB/OL]. [2025-10-02]. https://doi.org/10.7303/syn3193805.

[25] Kavur A E, Gezer N S, Baris M, et al.. CHAOS challenge-combined (CT-MR) healthy abdominal organ segmentation. Medical image analysis, 2021, 69: 101950 J]

[26] WANG T, ZHAO X, ZHUANG L, et al. pFedSAM: personalized federated learning of segment anything model for medical image segmentation[EB/OL]. (2025-09-19) [2025-10-02]. https://arxiv.org/abs/2509.15638.

RIGHTS & PERMISSIONS

Tianjin University of Technology