借助 FLIP 让跨洲联邦学习可重复开展：一项多应用研究
英文标题：Making Cross-Continental Federated Learning Repeatable with FLIP: a Multi-Application Study
研究团队提出开源的 FLIP（Federated Learning Interoperability Platform，联邦学习互操作平台），旨在将医疗领域联邦学习的训练与评估流程标准化、可重复化。FLIP 将队列查询、院内 PACS 的按需 DICOM 调取、站点级项目审批及可复用的联邦学习任务类型封装为可组合服务。研究人员在英国和泰国两个客户端节点的合成胸部 X 光队列上，完成了联邦微调与联邦评估两项验证。论文的核心主张并非比较联邦训练与集中式训练的算法优劣，而是证明该平台可提升国际协作中的可重复性、可审计性及站点自主治理能力。
2.1 KB
640 B
612 B
380 B
9.1 KB
32.7 KB
22.8 KB
24.2 KB
34.7 KB
中文
[提交于 2026 年 9 月 28 日]
标题：借助 FLIP 让跨洲联邦学习可重复开展：一项多应用研究
作者：Rafael Garcia-Dias、Alexandre Triay Bagur、Chayanin Tangwiriyasakul、Virginia Fernandez、Parhom Esmaeili、Piyalitt Ittichaiwong、Yang Li、Lawrence Adams、Wason Buncharoen、Martin Chapman、Benjamaporn Chayanond、Sadthavud Chunrod、Tanawat Fongsri、Kass Gibson、Supat Plungprasertkul、Supawit Tangpanithandee、Kanyakorn Veerakanjana、Vicky Goh、Michela Antonelli、Joe Zhang、Kongkiat Kespechara、Sebastien Ourselin、M. Jorge Cardoso
查看 Rafael Garcia-Dias 与其他 22 位作者撰写的论文《借助 FLIP 让跨洲联邦学习可重复开展：一项多应用研究》的 PDF。
摘要：医疗领域的联邦学习（FL）仍然面临挑战，因为每次合作都需要重新建立治理保障，其开销使大多数项目停留在概念验证阶段。本文提出 FLIP（Federated Learning Interoperability Platform，联邦学习互操作平台）：一个开源、多应用平台，使联邦学习训练和评估能够重复开展。FLIP 将常见的联邦学习工作流实现为一组可组合服务，包括针对各站点结构化数据库的队列查询、从机构 PACS 按需调取 DICOM、各站点的项目审批，以及可复用的联邦学习任务类型。为验证 FLIP，我们在位于英国（UK）和泰国的两个客户端节点上，使用合成胸部 X 光队列运行了两种不同用例：联邦微调和联邦评估。在 FLIP 中，每家机构均独立批准其参与每个项目，并按照本地 IT 安全流程运营自己的节点。本研究提出的是一项运营层面而非算法层面的主张。本文不比较联邦训练与集中式训练；针对这一问题，读者可参考已有的系统综述和 Meta 分析。核心结果表明，这类平台能够支持国际联邦学习协作，并提高可重复性、可审计性和站点特定的治理能力。我们还对现有平台进行了全面比较，以帮助研究人员和运营人员针对其用例选择合适的平台。
提交历史
来自：Rafael Garcia-Dias [
]
[v1]
2026 年 9 月 28 日，周一，18:00:07 UTC（1,600 KB）
English / 原文
[Submitted on 28 Sep 2026]
Title:Making Cross-Continental Federated Learning Repeatable with FLIP: a Multi-Application Study
Authors:Rafael Garcia-Dias
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
View a PDF of the paper titled Making Cross-Continental Federated Learning Repeatable with FLIP: a Multi-Application Study, by Rafael Garcia-Dias and 22 other authors
Abstract:Federated learning (FL) in healthcare remains challenging, as the overhead of rebuilding governance guarantees for every collaboration stops most projects at the proof-of-concept stage. Here we present FLIP (Federated Learning Interoperability Platform), an open-source, multi-application platform that makes FL training and evaluation repeatable. FLIP implements common FL workflows as a set of composable services: cohort queries against per-site structured databases, on-demand DICOM retrieval from institutional PACS, per-site project approval, and reusable FL job types. To demonstrate FLIP, we ran two distinct use cases, federated fine-tuning and federated evaluation, on synthetic chest X-ray cohorts across two client nodes based in the United Kingdom (UK) and Thailand. In FLIP, each institution independently approves its participation in each project and operates its own node under local IT security processes. This study makes an operational rather than an algorithmic claim. It does not compare federated with centralised training; for that question, we refer the reader to existing systematic reviews and meta-analyses. The central result is evidence that such platforms enable international FL collaboration and improve repeatability, auditability, and site-specific governance. We also present a comprehensive comparison of existing platforms to help researchers and operators choose the right platform for their use case.
Submission history
From: Rafael Garcia-Dias [
]
[v1]
Mon, 28 Sep 2026 18:00:07 UTC (1,600 KB)