---
title: "Pixel-wise Divide and Conquer for Federated Vessel Segmentation"
authors:
  - "Tian Chen"
  - "Wenke Huang"
  - "Zhihao Wang"
  - "Zekun Shi"
  - "He Li"
  - "Wenhui Dong"
  - "Mang Ye"
  - "Bo Du"
  - "Yongchao Xu"
doi: "10.24963/ijcai.2025/540"
url: "https://ijcai.org/proceedings/2025/540"
pdf: "https://ijcai.org/proceedings/2025/0540.pdf"
venue: "Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence"
year: 2025
pages: "4851-4859"
track: "Main Track"
keywords:
  - "Computer Vision: CV: Transparency, accountability, fairness and privacy"
  - "Computer Vision: CV: Biomedical image analysis"
  - "Machine Learning: ML: Federated learning"
  - "Computer Vision: CV: Segmentation, grouping and shape analysis"
---

# Pixel-wise Divide and Conquer for Federated Vessel Segmentation

Tian Chen, Wenke Huang, Zhihao Wang, Zekun Shi, He Li, Wenhui Dong, Mang Ye, Bo Du, Yongchao Xu

*Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence*, pages 4851-4859. DOI: [10.24963/ijcai.2025/540](https://doi.org/10.24963/ijcai.2025/540)

## Abstract

Accurate vessel segmentation is essential for diagnosing and managing vascular and ophthalmic diseases. Traditional learning-based vessel segmentation methods heavily rely on high-quality, pixel-level annotated datasets. However, segmentation performance suffers significantly when applied in federated learning settings due to vessel morphology inconsistency and vessel-background imbalance. The former limits the ability of models to capture fine-grained vessels, while the latter overemphasizes background pixels and biases the model towards them. To address these challenges, we propose a novel method named Federated Vessel-Aware Calibration (FVAC),  which leverages global uncertainty to provide differentiated guidance for clients, focusing on pixels of various morphologies that are difficult to distinguish. Furthermore, we introduce a foreground-background decoupling alignment strategy that utilizes more stable and balanced global features to mitigate semantic drift caused by vessel-background imbalance in local clients. Comprehensive experiments confirm the effectiveness of our method