mlsanityData to 14 Oct 2026

Decouple, Purify and Unite: Semantic-Structural Prototype Learning for Federated Medical Segmentation

arXiv 2610.04700 ↗First posted 3 Oct 2026cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Federated learning enables medical institutions to train a global model without sharing data, yet feature heterogeneity from diverse scanners or protocols remains challenging. Existing representation-based methods face two limitations: 1) Incomplete Contextual Representation Learning: single-layer or coupled representations overlook multi-level structural cues and entangle regional semantics with boundary details. 2) Layerwise Style and Aggregation Biases: domain-specific style discrepancies across intermediate layers degrade prototypes, while aggregation that overlooks client distribution shifts can further amplify bias. We propose FedBCS+, federated decoupled contextual alignment with style-purified aggregation. We employ Frequency-domain Style Recalibration (FSR) in prototype construction to decouple content-style representations and extract style-purified prototypes. Built upon these purified features, Decoupled Contextual Prototype Alignment (DCPA) explicitly decouples multi-level features into semantic and structural prototypes and aligns regional semantics and fine-grained anatomical structures separately. Style-purified Semantic Prototype Aggregation (S2PA) measures each client's purified prototype divergence from the global consensus and adaptively reweights aggregation toward under-represented clients to reduce consensus bias. On five heterogeneous medical segmentation benchmarks spanning histopathology, MRI, ultrasound, and colonoscopy, FedBCS+ achieves the highest mean Dice among the compared methods. A convergence analysis further characterizes how aggregation and alignment affect the optimization bound.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 43 of its 72 references that are in the map. First posted on arXiv 3 Oct 2026.

Builds on

43 of 72 references.

References · 43

  1. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  2. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,669
  3. 2018Federated Optimization in Heterogeneous Networkscited by 2,987
  4. 2017Arbitrary Style Transfer in Real-time with Adaptive Instance Normalizationcited by 2,010
  5. 2016Instance Normalization: The Missing Ingredient for Fast Stylizationcited by 1,198
  6. 2020Tackling the Objective Inconsistency Problem in Heterogeneous Federated Optimizationcited by 906
  7. 2021Model-Contrastive Federated Learningcited by 766
  8. 2020Ensemble Distillation for Robust Model Fusion in Federated Learningcited by 682
  9. 2020Ditto: Fair and Robust Federated Learning Through Personalizationcited by 672
  10. 2021FedBN: Federated Learning on Non-IID Features via Local Batch Normalizationcited by 571
  11. 2021Federated Learning on Non-IID Data Silos: An Experimental Studycited by 568
  12. 2016Federated Optimization: Distributed Machine Learning for On-Device Intelligencecited by 564
31 more
  1. 2021FedProto: Federated Prototype Learning across Heterogeneous Clientscited by 352
  2. 2021A Fourier-based Framework for Domain Generalizationcited by 290
  3. 2021FedDG: Federated Domain Generalization on Medical Image Segmentation via Episodic Learning in Continuous Frequency Spacecited by 287
  4. 2016A Benchmark for Endoluminal Scene Segmentation of Colonoscopy Imagescited by 221
  5. 2020Medical Image Segmentation Using Deep Learning: A Surveycited by 164
  6. 2023Heterogeneous Federated Learning: State-of-the-art and Research Challengescited by 161
  7. 2021FairFed: Enabling Group Fairness in Federated Learningcited by 118
  8. 2020Improving Semantic Segmentation via Decoupled Body and Edge Supervisioncited by 99
  9. 2023Federated Learning for Generalization, Robustness, Fairness: A Survey and Benchmarkcited by 99
  10. 2019Efficient Parameter-free Clustering Using First Neighbor Relationscited by 97
  11. 2021Amplitude-Phase Recombination: Rethinking Robustness of Convolutional Neural Networks in Frequency Domaincited by 71
  12. 2022Tackling Data Heterogeneity in Federated Learning with Class Prototypescited by 67
  13. 2021HarmoFL: Harmonizing Local and Global Drifts in Federated Learning on Heterogeneous Medical Imagescited by 64
  14. 2023Fair Federated Medical Image Segmentation via Client Contribution Estimationcited by 52
  15. 2021The Federated Tumor Segmentation (FeTS) Challengecited by 50
  16. 2023Generalizable Heterogeneous Federated Cross-Correlation and Instance Similarity Learningcited by 40
  17. 2024Fair Federated Learning under Domain Skew with Local Consistency and Domain Diversitycited by 30
  18. 2022Personalizing Federated Medical Image Segmentation via Local Calibrationcited by 29
  19. 2023Think Twice Before Selection: Federated Evidential Active Learning for Medical Image Analysis with Domain Shiftscited by 28
  20. 2024Taming Cross-Domain Representation Variance in Federated Prototype Learning with Heterogeneous Data Domainscited by 28
  21. 2023Decompose, Adjust, Compose: Effective Normalization by Playing with Frequency for Domain Generalizationcited by 26
  22. 2022DecoupleNet: Decoupled Network for Domain Adaptive Semantic Segmentationcited by 23
  23. 2020Dataset of Segmented Nuclei in Hematoxylin and Eosin Stained Histopathology Images of 10 Cancer Typescited by 17
  24. 2023A Simple Data Augmentation for Feature Distribution Skewed Federated Learningcited by 17
  25. 2024Rethinking the Representation in Federated Unsupervised Learning with Non-IID Datacited by 13
  26. 2025Geometric Knowledge-Guided Localized Global Distribution Alignment for Federated Learningcited by 12
  27. 2024FedUV: Uniformity and Variance for Heterogeneous Federated Learningcited by 9
  28. 2024MoreStyle: Relax Low-frequency Constraint of Fourier-based Image Reconstruction in Generalizable Medical Image Segmentationcited by 9
  29. 2025PathFL: Multi-Alignment Federated Learning for Pathology Image Segmentationcited by 3
  30. 2025Divide, Conquer and Unite: Hierarchical Style-Recalibrated Prototype Alignment for Federated Medical Segmentationcited by 3
  31. 2025pFedSAM: Personalized Federated Learning of Segment Anything Model for Medical Image Segmentationcited by 1

Built on by

Not cited yet.