FedOnco-Bench: A Reproducible Benchmark for Privacy-Aware Federated Tumor Segmentation with Synthetic CT Data

2025-11-02

AI Summary

This work addresses the dual challenges of membership inference attacks and data heterogeneity in federated learning for medical image segmentation. We propose the first open-source, privacy-aware benchmark framework specifically designed for tumor segmentation. Built upon synthetic CT images, it provides a reproducible and standardized evaluation platform enabling multi-institutional collaborative training while preserving data privacy. The framework integrates mainstream federated algorithms—including FedAvg, FedProx, FedBN, and DP-SGD—and systematically characterizes the privacy–utility trade-off. Empirical results show that DP-SGD reduces membership inference attack AUC to 0.25, substantially enhancing privacy protection; meanwhile, FedProx and FedBN achieve superior segmentation accuracy under non-IID data conditions. This work establishes a scalable, empirically grounded evaluation paradigm for privacy-preserving federated learning in medical imaging.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Computer Vision: Segmentation
• Natural Language Processing: Safety and Robustness

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Security and Privacy: Data transparency and provenance
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing

Abstract

Federated Learning (FL) allows multiple institutions to cooperatively train machine learning models while retaining sensitive data at the source, which has great utility in privacy-sensitive environments. However, FL systems remain vulnerable to membership-inference attacks and data heterogeneity. This paper presents FedOnco-Bench, a reproducible benchmark for privacy-aware FL using synthetic oncologic CT scans with tumor annotations. It evaluates segmentation performance and privacy leakage across FL methods: FedAvg, FedProx, FedBN, and FedAvg with DP-SGD. Results show a distinct trade-off between privacy and utility: FedAvg is high performance (Dice around 0.85) with more privacy leakage (attack AUC about 0.72), while DP-SGD provides a higher level of privacy (AUC around 0.25) at the cost of accuracy (Dice about 0.79). FedProx and FedBN offer balanced performance under heterogeneous data, especially with non-identical distributed client data. FedOnco-Bench serves as a standardized, open-source platform for benchmarking and developing privacy-preserving FL methods for medical image segmentation.

Problem

• Benchmarking privacy-aware federated tumor segmentation methods
• Evaluating privacy-utility trade-offs in medical image analysis
• Addressing data heterogeneity in federated learning for oncology

Innovation

• Developed privacy-aware federated tumor segmentation benchmark
• Evaluated federated learning methods with synthetic CT data
• Assessed privacy-utility trade-off across different FL approaches