pith. sign in

arxiv: 2606.23500 · v1 · pith:NQIQ3H6Lnew · submitted 2026-06-22 · 💻 cs.DC · cs.LG

Development and Design of FLKit: A Structured Onboarding Toolkit for Federated Learning in Health and Life Sciences

Ashkan Pirmani , Ilse Vermeulen , Goran Vinterhalter , Lotte Geys , Axel Faes , Muhammad Quamber Ali , Nishkala Sattanathan , Geert Vandeweyer
show 2 more authors
Yves Moreau Liesbet M. Peeters
This is my paper

Pith reviewed 2026-06-26 07:19 UTC · model grok-4.3

classification 💻 cs.DC cs.LG
keywords federated learningonboarding toolkithealth and life sciencesmultidisciplinary teamsgovernanceinfrastructureFAIR principlescommunity maintained
0
0 comments X

The pith

FLKit supplies an open toolkit with role-aware pathways for starting federated learning projects in health and life sciences.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper describes the creation of FLKit to address the practical barriers teams face when beginning federated learning projects, where frameworks, governance rules, and roles are scattered. It positions FLKit as a community-maintained resource modeled after the ELIXIR Research Data Management Kit. The toolkit organizes content around four stages of the lifecycle and provides eleven role-specific entry points along with supporting materials like a glossary and project template. A sympathetic reader would care because it offers a concrete way for diverse contributors to collaborate without each needing expertise in all areas. Since its demo, it has expanded with documented project stories from real applications in multiple sclerosis, inflammatory bowel disease, genomics, and brain-computer interfaces.

Core claim

FLKit is an open, community-maintained onboarding toolkit that takes a multidisciplinary team through the full federated learning lifecycle and gives every contributor a role-aware entry point. It is built on four lifecycle stages—Governance, Infrastructure, Wrangling, and Analysis—connected by eleven role-specific entry points, a cross-disciplinary glossary, a reusable FAIR-aligned FL Story template, and a curated directory of tools. The content was developed with a multidisciplinary core team, consortium milestone reviews, and external practitioner interviews, and has grown to 39 pages with seven documented FL Stories in areas such as multiple sclerosis and genomics.

What carries the argument

The FLKit toolkit, structured around four lifecycle stages and eleven role-specific entry points that organize onboarding content for clinical, legal, governance, and technical roles.

If this is right

  • Multidisciplinary teams gain a single starting point instead of piecing together scattered resources.
  • Projects can use the FL Story template to plan and document in a FAIR-aligned way.
  • The toolkit remains open for community contributions to keep it current with new tools and regulations.
  • Real projects in multiple sclerosis disability prediction, inflammatory bowel disease, genomics, and brain-computer interfaces have already used the structure.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Similar role-based onboarding structures could help in other regulated data domains such as finance.
  • The emphasis on community maintenance suggests the toolkit will evolve as new frameworks and privacy rules appear.
  • Tracking which sections different roles consult most often could show where the design succeeds or needs adjustment.

Load-bearing premise

That input from a multidisciplinary core team, wider consortium milestone reviews, and external practitioner interviews is sufficient to ground the toolkit content in real-world practice and make it effective for diverse institutions starting federated projects.

What would settle it

A comparison of project start times and coordination issues between teams that follow FLKit versus similar teams that do not would test whether the role-aware structure reduces the described barriers.

read the original abstract

Federated learning lets institutions train shared models without moving their data, which makes it a natural fit for health and life sciences research under strict privacy regulation. The methods are maturing fast, but the practical barrier now comes earlier: a team starting a federated project meets a scattered mix of frameworks, governance obligations, and unfamiliar roles, with no structured place to begin that fits its own background. FLKit closes that gap. It is an open, community-maintained onboarding toolkit that takes a multidisciplinary team through the full federated learning lifecycle and gives every contributor, clinical, legal, governance, or technical, a role-aware entry point instead of assuming fluency across all four. We modeled it on the ELIXIR Research Data Management Kit and built it with a multidisciplinary core team, a wider consortium supplying milestone reviews and roadmap direction, and external practitioners interviewed to keep the content grounded in real practice. FLKit sits on four lifecycle stages, Governance, Infrastructure, Wrangling, and Analysis, and connects them through 11 role-specific entry points, a cross-disciplinary glossary, a reusable FAIR-aligned FL Story template for planning and documenting projects, and a curated directory of tools, frameworks, and communities. Since the December 2024 demo it has grown to 39 pages across eight sections, with seven FL Stories documenting completed and ongoing projects in multiple sclerosis disability prediction, inflammatory bowel disease, genomics, and brain-computer interfaces. It is openly available at https://uhasselt-biomedicaldatasciences.github.io/federated-learning-toolkit/ and welcomes contributions from across the life sciences.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 1 minor

Summary. The manuscript presents FLKit, an open, community-maintained onboarding toolkit for federated learning in health and life sciences. It claims to close the practical barrier gap for multidisciplinary teams by offering a structured approach through four lifecycle stages (Governance, Infrastructure, Wrangling, Analysis), 11 role-specific entry points, a cross-disciplinary glossary, a reusable FAIR-aligned FL Story template, and a curated directory of tools and communities. The toolkit was developed with input from a multidisciplinary core team, wider consortium milestone reviews and roadmap direction, and external practitioner interviews; since its December 2024 demo it has grown to 39 pages across eight sections with seven FL Stories documenting projects in multiple sclerosis, inflammatory bowel disease, genomics, and brain-computer interfaces. It is available at https://uhasselt-biomedicaldatasciences.github.io/federated-learning-toolkit/.

Significance. If validated through usage data, FLKit could have meaningful significance by lowering entry barriers for federated learning projects in privacy-regulated domains, improving role coordination across clinical, legal, governance, and technical contributors, and providing a reusable template and resource directory that accelerates project planning. The open, community-driven model and explicit modeling on the ELIXIR Research Data Management Kit are constructive elements that could support wider adoption and iterative improvement.

major comments (1)
  1. [Abstract] Abstract: The assertion that FLKit 'closes that gap' by supplying role-aware entry points and taking teams through the full lifecycle is unsupported. The manuscript supplies only a narrative of the development process plus descriptive growth metrics (39 pages, 7 FL Stories) and contains no evaluation data, user testing results, adoption metrics, pre/post assessments, or controlled feedback showing measurable reductions in onboarding barriers or improvements in project initiation, role coordination, or error avoidance.
minor comments (1)
  1. [Abstract] Abstract: The abstract is lengthy and interleaves descriptive content with impact claims; a more concise structure separating the toolkit description from the effectiveness assertions would improve clarity.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their constructive review and positive assessment of FLKit's potential. We respond to the single major comment below.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The assertion that FLKit 'closes that gap' by supplying role-aware entry points and taking teams through the full lifecycle is unsupported. The manuscript supplies only a narrative of the development process plus descriptive growth metrics (39 pages, 7 FL Stories) and contains no evaluation data, user testing results, adoption metrics, pre/post assessments, or controlled feedback showing measurable reductions in onboarding barriers or improvements in project initiation, role coordination, or error avoidance.

    Authors: We agree that the manuscript presents no formal evaluation data, user testing results, or controlled metrics, as it is a design and development paper describing the toolkit's structure, creation process, and initial content rather than an empirical study of its effectiveness. The abstract's phrasing draws from the documented development steps (multidisciplinary core team, consortium input, and practitioner interviews) but does not claim measured outcomes. To address the concern directly, we will revise the abstract to change 'FLKit closes that gap' to 'FLKit is designed to address this gap' and add a clarifying sentence that empirical validation of onboarding impact is planned as future work. This revision will appear in the next manuscript version. revision: yes

Circularity Check

0 steps flagged

No circularity; purely descriptive toolkit paper with no derivations or self-referential reductions

full rationale

The manuscript contains no equations, predictions, fitted parameters, or derivation chains. It narrates the construction of FLKit from a multidisciplinary process and external inputs (ELIXIR model, consortium reviews, interviews) without claiming that any result follows by construction from those inputs or from self-citations. All listed circularity patterns are absent; the central claim is a design description, not a mathematical or predictive reduction.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

This paper describes the development of a software toolkit rather than a scientific model or derivation, so there are no free parameters, axioms, or invented entities in the mathematical sense.

pith-pipeline@v0.9.1-grok · 5867 in / 1220 out tokens · 31995 ms · 2026-06-26T07:19:25.363197+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

31 extracted references · 24 canonical work pages · 2 internal anchors

  1. [1]

    Original Paper Development and Design of FLKit: A Structured Onboarding Toolkit for Federated Learning in Health and Life Sciences Ashkan Pirmani*,1,2,3, Ilse Vermeulen*,2,3, Goran Vinterhalter1, Lotte Geys2,3, Axel Faes2,3, Muhammad Quamber Ali1, Nishkala Sattanathan4, Geert Vandeweyer4, Yves Moreau1, Liesbet M. Peeters2,3 1STADIUS, Department of Electri...

    2024

  2. [2]

    the user

    to federated settings have raised awareness of data governance requirements, but these principles do not yet translate into practical onboarding for the full range of roles involved. The result is a fragmented entry experience: technical documentation, legal frameworks, and community resources all exist, but they are scattered across disconnected communit...

    2025

  3. [3]

    The future of digital health with federated learning

    Rieke N, Hancox J, Li W, Milletarì F, Roth HR, Albarqouni S, et al. The future of digital health with federated learning. Npj Digit Med. 2020 Sep 14;3(1):119. doi:10.1038/s41746-020-00323-1

    work page doi:10.1038/s41746-020-00323-1 2020

  4. [4]

    General Data Protection Regulation (GDPR). Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data. Off J Eur Union. 2016;L119:1-88. Available from: https://eur-lex.europa.eu/eli/reg/2016/679/oj

    2016

  5. [5]

    Public Law 104-191, 110 Stat

    Health Insurance Portability and Accountability Act of 1996 (HIPAA). Public Law 104-191, 110 Stat

    1996

  6. [6]

    Available from: http://arxiv.org/abs/1602.05629 doi:10.48550/arXiv.1602.05629

    work page doi:10.48550/arxiv.1602.05629

  7. [7]

    Secure, privacy-preserving and federated machine learning in medical imaging

    Kaissis GA, Makowski MR, Rückert D, Braren RF. Secure, privacy-preserving and federated machine learning in medical imaging. Nat Mach Intell. 2020 Jun 8;2(6):305-11. doi:10.1038/s42256-020-0186-1

    work page doi:10.1038/s42256-020-0186-1 2020

  8. [8]

    Federated Learning for Smart Healthcare: A Survey

    Nguyen DC, Pham QV, Pathirana PN, Ding M, Seneviratne A, Lin Z, et al. Federated Learning for Smart Healthcare: A Survey. ACM Comput Surv. 2023 Mar 31;55(3):1-37. doi:10.1145/3501296

    work page doi:10.1145/3501296 2023

  9. [9]

    Federated machine learning in healthcare: A systematic review on clinical applications and technical architecture

    Teo ZL, Jin L, Liu N, Li S, Miao D, Zhang X, et al. Federated machine learning in healthcare: A systematic review on clinical applications and technical architecture. Cell Rep Med. 2024 Feb;5(2):101419. doi:10.1016/j.xcrm.2024.101419

    work page doi:10.1016/j.xcrm.2024.101419 2024

  10. [10]

    Federated Learning for Healthcare: Systematic Review and Architecture Proposal

    Antunes RS, André Da Costa C, Küderle A, Yari IA, Eskofier B. Federated Learning for Healthcare: Systematic Review and Architecture Proposal. ACM Trans Intell Syst Technol. 2022 Aug 31;13(4):1-23. doi:10.1145/3501813

    work page doi:10.1145/3501813 2022

  11. [11]

    Available from: http://arxiv.org/abs/2007.14390 doi:10.48550/arXiv.2007.14390

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2007.14390 2007

  12. [12]

    p. 201-10. (Lecture Notes in Computer Science). doi:10.1007/978-3-030-60548-3_20

    work page doi:10.1007/978-3-030-60548-3_20

  13. [13]

    Available from: https://arxiv.org/abs/1811.04017 doi:10.48550/arXiv.1811.04017

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1811.04017

  14. [14]

    Deep Learning with Differential Privacy

    Abadi M, Chu A, Goodfellow I, McMahan HB, Mironov I, Talwar K, et al. Deep Learning with Differential Privacy. In: Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security. Vienna, Austria: ACM

    2016

  15. [15]

    p. 308-18. doi:10.1145/2976749.2978318

    work page doi:10.1145/2976749.2978318

  16. [16]

    Federated Learning With Differential Privacy: Algorithms and Performance Analysis

    Wei K, Li J, Ding M, Ma C, Yang HH, Farokhi F, et al. Federated Learning With Differential Privacy: Algorithms and Performance Analysis. IEEE Trans Inf Forensics Secur. 2020;15:3454-69. doi:10.1109/TIFS.2020.2988575

    work page doi:10.1109/tifs.2020.2988575 2020

  17. [17]

    Practical Secure Aggregation for Privacy-Preserving Machine Learning

    Bonawitz K, Ivanov V, Kreuter B, Marcedone A, McMahan HB, Patel S, et al. Practical Secure Aggregation for Privacy-Preserving Machine Learning. In: Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security. Dallas, Texas, USA: ACM

    2017

  18. [18]

    p. 1175-91. doi:10.1145/3133956.3133982

    work page doi:10.1145/3133956.3133982

  19. [19]

    Federated Learning for Privacy Preservation in Smart Healthcare Systems: A Comprehensive Survey

    Ali M, Naeem F, Tariq M, Kaddoum G. Federated Learning for Privacy Preservation in Smart Healthcare Systems: A Comprehensive Survey. IEEE J Biomed Health Inform. 2023 Feb;27(2):778-89. doi:10.1109/JBHI.2022.3181823

    work page doi:10.1109/jbhi.2022.3181823 2023

  20. [20]

    Federated electronic health records for the European Health Data Space

    Raab R, Küderle A, Zakreuskaya A, Stern AD, Klucken J, Kaissis G, et al. Federated electronic health records for the European Health Data Space. Lancet Digit Health. 2023 Nov;5(11):e840-7. doi:10.1016/S2589-7500(23)00156-5

    work page doi:10.1016/s2589-7500(23)00156-5 2023

  21. [21]

    The European health data space: Too big to succeed? Health Policy

    Marelli L, Stevens M, Sharon T, Van Hoyweghen I, Boeckhout M, Colussi I, et al. The European health data space: Too big to succeed? Health Policy. 2023 Sep;135:104861. doi:10.1016/j.healthpol.2023.104861

    work page doi:10.1016/j.healthpol.2023.104861 2023

  22. [22]

    Scientific data3, 160018:1–9 (2016).https://doi.org/10.1038/sdata.2016.18

    Wilkinson MD, Dumontier M, Aalbersberg IjJ, Appleton G, Axton M, Baak A, et al. The FAIR Guiding Principles for scientific data management and stewardship. Sci Data. 2016 Mar 15;3(1):160018. doi:10.1038/sdata.2016.18

    work page doi:10.1038/sdata.2016.18 2016

  23. [23]

    Feasibility and utility of applications of the common data model to multiple, disparate observational health databases

    Voss EA, Makadia R, Matcho A, Ma Q, Knoll C, Schuemie M, et al. Feasibility and utility of applications of the common data model to multiple, disparate observational health databases. J Am Med Inform Assoc. 2015 May 1;22(3):553-64. doi:10.1093/jamia/ocu023

    work page doi:10.1093/jamia/ocu023 2015

  24. [24]

    Federated Machine Learning, Privacy-Enhancing Technologies, and Data Protection Laws in Medical Research: Scoping Review

    Brauneck A, Schmalhorst L, Kazemi Majdabadi MM, Bakhtiari M, Völker U, Baumbach J, et al. Federated Machine Learning, Privacy-Enhancing Technologies, and Data Protection Laws in Medical Research: Scoping Review. J Med Internet Res. 2023 Mar 30;25:e41588. doi:10.2196/41588

    work page doi:10.2196/41588 2023

  25. [25]

    URLhttps://doi.org/10.1016/j.cels.2025

    Alper P, D'Anna F, Droesbeke B, Andrabi M, Andrade Buono R, Bianchini F, et al. RDMkit: A research data management toolkit for life sciences. Patterns. 2025 Sep;6(9):101345. doi:10.1016/j.patter.2025.101345

    work page doi:10.1016/j.patter.2025.101345 2025

  26. [26]

    Accessible Ecosystem for Clinical Research (Federated Learning for Everyone): Development and Usability Study

    Pirmani A, Oldenhof M, Peeters LM, De Brouwer E, Moreau Y. Accessible Ecosystem for Clinical Research (Federated Learning for Everyone): Development and Usability Study. JMIR Form Res. 2024 Jul 17;8:e55496. doi:10.2196/55496

    work page doi:10.2196/55496 2024

  27. [27]

    The Journey of Data Within a Global Data Sharing Initiative: A Federated 3-Layer Data Analysis Pipeline to Scale Up Multiple Sclerosis Research

    Pirmani A, De Brouwer E, Geys L, Parciak T, Moreau Y, Peeters LM. The Journey of Data Within a Global Data Sharing Initiative: A Federated 3-Layer Data Analysis Pipeline to Scale Up Multiple Sclerosis Research. JMIR Med Inform. 2023 Nov 9;11:e48030. doi:10.2196/48030

    work page doi:10.2196/48030 2023

  28. [28]

    Personalized federated learning for predicting disability progression in multiple sclerosis using real-world routine clinical data

    Pirmani A, De Brouwer E, Arany Á, Oldenhof M, Passemiers A, Faes A, et al. Personalized federated learning for predicting disability progression in multiple sclerosis using real-world routine clinical data. Npj Digit Med. 2025 Jul 24;8(1):478. doi:10.1038/s41746-025-01788-8

    work page doi:10.1038/s41746-025-01788-8 2025

  29. [29]

    WiNGS-API: a federated genome/phenome data sharing platform enabling gene discovery and variant classification for rare diseases

    Sattanathan N, Huremagic B, Vermeesch JR, Moreau Y, Vandeweyer G. WiNGS-API: a federated genome/phenome data sharing platform enabling gene discovery and variant classification for rare diseases. Genome Med. 2026 Mar 26;18(1):37. doi:10.1186/s13073-026-01627-9

    work page doi:10.1186/s13073-026-01627-9 2026

  30. [30]

    Available from: https://arxiv.org/abs/2412.06815 doi:10.48550/arXiv.2412.06815

    work page doi:10.48550/arxiv.2412.06815

  31. [31]

    https://faircookbook.elixir-europe.org/content/home.html, [Accessed 09-06-2026]

    FAIR Cookbook --- faircookbook.elixir-europe.org. https://faircookbook.elixir-europe.org/content/home.html, [Accessed 09-06-2026]

    2026