Collaborative and privacy-preserving cross-vendor united diagnostic imaging via server-rotating federated machine learning

Abstract Federated Learning (FL) is a distributed framework that enables collaborative training of a server model across medical data vendors while preserving data privacy. However, conventional FL faces two key challenges: substantial data heterogeneity among vendors and limited flexibility from a...

Full description

Saved in:
Bibliographic Details
Main Authors: Hao Wang, Xiaoyu Zhang, Xuebin Ren, Zheng Zhang, Shusen Yang, Chunfeng Lian, Jianhua Ma, Dong Zeng
Format: Article
Language:English
Published: Nature Portfolio 2025-08-01
Series:Communications Engineering
Online Access:https://doi.org/10.1038/s44172-025-00485-4
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849763473565679616
author Hao Wang
Xiaoyu Zhang
Xuebin Ren
Zheng Zhang
Shusen Yang
Chunfeng Lian
Jianhua Ma
Dong Zeng
author_facet Hao Wang
Xiaoyu Zhang
Xuebin Ren
Zheng Zhang
Shusen Yang
Chunfeng Lian
Jianhua Ma
Dong Zeng
author_sort Hao Wang
collection DOAJ
description Abstract Federated Learning (FL) is a distributed framework that enables collaborative training of a server model across medical data vendors while preserving data privacy. However, conventional FL faces two key challenges: substantial data heterogeneity among vendors and limited flexibility from a fixed server, leading to suboptimal performance in diagnostic-imaging tasks. To address these, we propose a server-rotating federated learning method (SRFLM). Unlike traditional FL, SRFLM designates one vendor as a provisional server for federated fine-tuning, with others acting as clients. It uses a rotational server-communication mechanism and a dynamic server-election strategy, allowing each vendor to sequentially assume the server role over time. Additionally, the communication protocol of SRFLM provides strong privacy guarantees using differential privacy. We extensively evaluate SRFLM across multiple cross-vendor diagnostic imaging tasks. We envision SRFLM as paving the way to facilitate collaborative model training across medical data vendors, thereby achieving the goal of cross-vendor united diagnostic imaging.
format Article
id doaj-art-57cc780fac064a7a8b8e107868cffd9a
institution DOAJ
issn 2731-3395
language English
publishDate 2025-08-01
publisher Nature Portfolio
record_format Article
series Communications Engineering
spelling doaj-art-57cc780fac064a7a8b8e107868cffd9a2025-08-20T03:05:24ZengNature PortfolioCommunications Engineering2731-33952025-08-014111610.1038/s44172-025-00485-4Collaborative and privacy-preserving cross-vendor united diagnostic imaging via server-rotating federated machine learningHao Wang0Xiaoyu Zhang1Xuebin Ren2Zheng Zhang3Shusen Yang4Chunfeng Lian5Jianhua Ma6Dong Zeng7School of Biomedical Engineering, Southern Medical UniversitySchool of Biomedical Engineering, Southern Medical UniversityNational Engineering Laboratory for Big Data Analytics, Xi’an Jiaotong UniversityNational Engineering Laboratory for Big Data Analytics, Xi’an Jiaotong UniversityNational Engineering Laboratory for Big Data Analytics, Xi’an Jiaotong UniversityMinistry of Education Key Laboratory for Intelligent Networks and Network Security, Xi’an Jiaotong UniversitySchool of Biomedical Engineering, Southern Medical UniversitySchool of Biomedical Engineering, Southern Medical UniversityAbstract Federated Learning (FL) is a distributed framework that enables collaborative training of a server model across medical data vendors while preserving data privacy. However, conventional FL faces two key challenges: substantial data heterogeneity among vendors and limited flexibility from a fixed server, leading to suboptimal performance in diagnostic-imaging tasks. To address these, we propose a server-rotating federated learning method (SRFLM). Unlike traditional FL, SRFLM designates one vendor as a provisional server for federated fine-tuning, with others acting as clients. It uses a rotational server-communication mechanism and a dynamic server-election strategy, allowing each vendor to sequentially assume the server role over time. Additionally, the communication protocol of SRFLM provides strong privacy guarantees using differential privacy. We extensively evaluate SRFLM across multiple cross-vendor diagnostic imaging tasks. We envision SRFLM as paving the way to facilitate collaborative model training across medical data vendors, thereby achieving the goal of cross-vendor united diagnostic imaging.https://doi.org/10.1038/s44172-025-00485-4
spellingShingle Hao Wang
Xiaoyu Zhang
Xuebin Ren
Zheng Zhang
Shusen Yang
Chunfeng Lian
Jianhua Ma
Dong Zeng
Collaborative and privacy-preserving cross-vendor united diagnostic imaging via server-rotating federated machine learning
Communications Engineering
title Collaborative and privacy-preserving cross-vendor united diagnostic imaging via server-rotating federated machine learning
title_full Collaborative and privacy-preserving cross-vendor united diagnostic imaging via server-rotating federated machine learning
title_fullStr Collaborative and privacy-preserving cross-vendor united diagnostic imaging via server-rotating federated machine learning
title_full_unstemmed Collaborative and privacy-preserving cross-vendor united diagnostic imaging via server-rotating federated machine learning
title_short Collaborative and privacy-preserving cross-vendor united diagnostic imaging via server-rotating federated machine learning
title_sort collaborative and privacy preserving cross vendor united diagnostic imaging via server rotating federated machine learning
url https://doi.org/10.1038/s44172-025-00485-4
work_keys_str_mv AT haowang collaborativeandprivacypreservingcrossvendoruniteddiagnosticimagingviaserverrotatingfederatedmachinelearning
AT xiaoyuzhang collaborativeandprivacypreservingcrossvendoruniteddiagnosticimagingviaserverrotatingfederatedmachinelearning
AT xuebinren collaborativeandprivacypreservingcrossvendoruniteddiagnosticimagingviaserverrotatingfederatedmachinelearning
AT zhengzhang collaborativeandprivacypreservingcrossvendoruniteddiagnosticimagingviaserverrotatingfederatedmachinelearning
AT shusenyang collaborativeandprivacypreservingcrossvendoruniteddiagnosticimagingviaserverrotatingfederatedmachinelearning
AT chunfenglian collaborativeandprivacypreservingcrossvendoruniteddiagnosticimagingviaserverrotatingfederatedmachinelearning
AT jianhuama collaborativeandprivacypreservingcrossvendoruniteddiagnosticimagingviaserverrotatingfederatedmachinelearning
AT dongzeng collaborativeandprivacypreservingcrossvendoruniteddiagnosticimagingviaserverrotatingfederatedmachinelearning