Federated learning enables big data for rare cancer boundary detection

Mostra el registre complet Registre parcial de l'ítem

• dc.contributor.author Pati, Sarthak
• dc.contributor.author Capellades Font, Jaume
• dc.contributor.author Bakas, Spyridon
• dc.date.accessioned 2023-04-19T11:41:29Z
• dc.date.available 2023-04-19T11:41:29Z
• dc.date.issued 2022
• dc.description.abstract Although machine learning (ML) has shown promise across disciplines, out-of-sample generalizability is concerning. This is currently addressed by sharing multi-site data, but such centralization is challenging/infeasible to scale due to various limitations. Federated ML (FL) provides an alternative paradigm for accurate and generalizable ML, by only sharing numerical model updates. Here we present the largest FL study to-date, involving data from 71 sites across 6 continents, to generate an automatic tumor boundary detector for the rare disease of glioblastoma, reporting the largest such dataset in the literature (n = 6, 314). We demonstrate a 33% delineation improvement for the surgically targetable tumor, and 23% for the complete tumor extent, over a publicly trained model. We anticipate our study to: 1) enable more healthcare studies informed by large diverse data, ensuring meaningful results for rare diseases and underrepresented populations, 2) facilitate further analyses for glioblastoma by releasing our consensus model, and 3) demonstrate the FL effectiveness at such scale and task-complexity as a paradigm shift for multi-site collaborations, alleviating the need for data-sharing.
• dc.description.sponsorship Research and main methodological developments reported in this publication were partly supported by the National Institutes of Health (NIH) under award numbers NIH/NCI:U01CA242871 (S. Bakas), NIH/NINDS:R01NS042645 (C. Davatzikos), NIH/NCI:U24CA189523 (C. Davatzikos), NIH/NCI:U24CA215109 (J. Saltz), NIH/NCI:U01CA248226 (P. Tiwari), NIH/NCI:P30CA51008 (Y. Gusev), NIH:R50CA211270 (M. Muzi), NIH/NCATS:UL1TR001433 (Y. Yuan), NIH/NIBIB:R21EB030209 (Y. Yuan), NIH/NCI:R37CA214955 (A. Rao), and NIH:R01CA233888 (A.L. Simpson). The authors would also like to acknowledge the following NIH funded awards for the multi-site clinical trial (NCT00884741, RTOG0825/ACRIN6686): U10CA21661, U10CA37422, U10CA180820, U10CA180794, U01CA176110, R01CA082500, CA079778, CA080098, CA180794, CA180820, CA180822, CA180868. Research reported in this publication was also partly supported by the National Science Foundation, under award numbers 2040532 (S. Baek), and 2040462 (B. Landman). Research reported in this publication was also supported by i) a research grant from Varian Medical Systems (Palo Alto, CA, USA) (Y.Yuan), (ii) the Ministry of Health of the Czech Republic (Grant Nr. NU21-08-00359) (M.Kerkovský and M.Kozubek), (iii) Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) Project-ID 404521405, SFB 1389, Work Package C02, and Priority Program 2177 “Radiomics: Next Generation of Biomedical Imaging” (KI 2410/1-1 ∣ MA 6340/18-1) (P. Vollmuth), (iv) DFG Project-ID B12, SFB 824 (B. Wiestler), (v) the Helmholtz Association (funding number ZT-I-OO1 4) (K. Maier-Hein), vi) the Dutch Cancer Society (KWF project number EMCR 2015-7859) (S.R. van der Voort), (vii) the Chilean National Agency for Research and Development (ANID-Basal FB0008 (AC3E) and FB210017 (CENIA)) (P. Guevara), viii) the Canada CIFAR AI Chairs Program (M. Vallières), (ix) Leeds Hospital Charity (Ref: 9RO1/1403) (S. Currie), (x) the Cancer Research UK funding for the Leeds Radiotherapy Research Centre of Excellence (RadNet) and the grant number C19942/A28832 (S. Currie), (xi) Medical Research Council (MRC) Doctoral Training Program in Precision Medicine (Award Reference No. 2096671) (J. Bernal), (xii) The European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (Grant Agreement No. 757173) (B.Glocker), (xiii) The UKRI London Medical Imaging & Artificial Intelligence Centre for Value-Based Healthcare (K. Kamnitsas), (xiv) Wellcome/Engineering and Physical Sciences Research Council (EPSRC) Center for Medical Engineering (WT 203148/Z/16/Z) (T.C. Booth), (xv) American Cancer Society Research Scholar Grant RSG-16-005-01 (A. Rao), (xvi) the Department of Defense (DOD) Peer Reviewed Cancer Research Program (PRCRP) W81XWH-18-1-0404, Dana Foundation David Mahoney Neuroimaging Program, the V Foundation Translational Research Award, Johnson & Johnson WiSTEM2D Award (P. Tiwari), (xvii) RSNA Research & Education Foundation under grant number RR2011 (E.Calabrese), (xviii) the National Research Fund of Luxembourg (FNR) (grant number: C20/BM/14646004/GLASS-LUX/Niclou) (S.P.Niclou), xix) EU Marie Curie FP7-PEOPLE-2012-ITN project TRANSACT (PITN-GA-2012-316679) and the Swiss National Science Foundation (project number 140958) (J. Slotboom), and (xx) CNPq 303808/2018-7 and FAPESP 2014/12236-1 (A. Xavier Falcão). The content of this publication is solely the responsibility of the authors and does not represent the official views of the NIH, the NSF, the RSNA R&E Foundation, or any of the additional funding bodies.
• dc.format.mimetype application/pdf
• dc.identifier.citation Pati S, Baid U, Edwards B, Sheller M, Wang SH, Reina GA, et al. Federated learning enables big data for rare cancer boundary detection. Nat Commun. 2022 Dec 5;13(1):7346. DOI: 10.1038/s41467-022-33407-5
• dc.identifier.doi http://dx.doi.org/10.1038/s41467-022-33407-5
• dc.identifier.issn 2041-1723
• dc.identifier.uri http://hdl.handle.net/10230/56500
• dc.language.iso eng
• dc.publisher Nature Research
• dc.relation.ispartof Nat Commun. 2022 Dec 5;13(1):7346
• dc.relation.projectID info:eu-repo/grantAgreement/EC/H2020/757173
• dc.rights © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
• dc.rights.accessRights info:eu-repo/semantics/openAccess
• dc.rights.uri http://creativecommons.org/licenses/by/4.0/
• dc.subject.keyword Biomedical engineering
• dc.subject.keyword CNS cancer
• dc.subject.keyword Computer science
• dc.subject.keyword Medical imaging
• dc.subject.keyword Medical research
• dc.title Federated learning enables big data for rare cancer boundary detection
• dc.type info:eu-repo/semantics/article
• dc.type.version info:eu-repo/semantics/publishedVersion