UltraFlwr -- An Efficient Federated Medical and Surgical Object Detection Framework

2025-03-19

AI Summary

To address multi-center data silos, scarce annotations, and constrained computational resources in medical edge scenarios, this paper proposes YOLO-PA, a novel federated training framework tailored for YOLO architectures. YOLO-PA introduces a partial aggregation strategy wherein clients upload only critical subnetwork parameters—reducing communication overhead by up to 83%. We establish the first federated benchmark for medical object detection and validate YOLO-PA on BCCD and m2cai16-tool-locations datasets: client models consistently outperform both isolated training and full-parameter aggregation baselines. The framework is lightweight and deployable on resource-constrained edge devices, ensuring cross-institutional privacy preservation while enabling collaborative model training. YOLO-PA thus delivers a practical, decentralized solution for clinical applications such as surgical instrument tracking and cell counting.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Computer Vision: Other Foundations of Computer Vision
• Multiagent Systems: Other Foundations of Multi Agent Systems

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Security and Privacy: Security and privacy of machine learning and AI applications

Abstract

Object detection shows promise for medical and surgical applications such as cell counting and tool tracking. However, its faces multiple real-world edge deployment challenges including limited high-quality annotated data, data sharing restrictions, and computational constraints. In this work, we introduce UltraFlwr, a framework for federated medical and surgical object detection. By leveraging Federated Learning (FL), UltraFlwr enables decentralized model training across multiple sites without sharing raw data. To further enhance UltraFlwr's efficiency, we propose YOLO-PA, a set of novel Partial Aggregation (PA) strategies specifically designed for YOLO models in FL. YOLO-PA significantly reduces communication overhead by up to 83% per round while maintaining performance comparable to Full Aggregation (FA) strategies. Our extensive experiments on BCCD and m2cai16-tool-locations datasets demonstrate that YOLO-PA not only provides better client models compared to client-wise centralized training and FA strategies, but also facilitates efficient training and deployment across resource-constrained edge devices. Further, we also establish one of the first benchmarks in federated medical and surgical object detection. This paper advances the feasibility of training and deploying detection models on the edge, making federated object detection more practical for time-critical and resource-constrained medical and surgical applications. UltraFlwr is publicly available at https://github.com/KCL-BMEIS/UltraFlwr.

Problem

• Addresses challenges in medical object detection deployment.
• Enables decentralized training without sharing raw data.
• Reduces communication overhead while maintaining model performance.

Innovation

• Federated Learning for decentralized medical object detection
• YOLO-PA reduces communication overhead by 83%
• Efficient training on resource-constrained edge devices