TY - JOUR
T1 - Systematic Review of Privacy-Preserving Distributed Machine Learning From Federated Databases in Health Care
AU - Zerka, Fadila
AU - Barakat, Samir
AU - Walsh, Sean
AU - Bogowicz, Marta
AU - Leijenaar, Ralph T. H.
AU - Jochems, Arthur
AU - Miraglio, Benjamin
AU - Townend, David
AU - Lambin, Philippe
N1 - Funding Information:
Supported by European Research Council advanced grant ERC-ADG-2015 Grant No. 694812, Hypoximmuno; the Dutch technology Foundation Stichting Technische Wetenschappen Grant No. P14-19 Radiomics STRaTegy, which is the applied science division of Dutch Research Council (De Nederlandse Organisatie voor Wetenschappelijk) the Technology Program of the Ministry of Economic Affairs; Small and Medium-Sized Enterprises Phase 2 RAIL Grant No. 673780; EUROSTARS, DART Grant No. E10116 and DECIDE Grant No. E11541; the European Program PREDICT ITN Grant No. 766276; Third Joint Transnational Call 2016 JTC2016 “CLEARLY” Grant No. UM 2017-
Funding Information:
Supported by European Research Council advanced grant ERC-ADG-2015 Grant No. 694812, Hypoximmuno; the Dutch technology Foundation Stichting Technische Wetenschappen Grant No. P14-19 Radiomics STRaTegy, which is the applied science division of Dutch Research Council (De Nederlandse Organisatie voor Wetenschappelijk) the Technology Program of the Ministry of Economic Affairs; Small and Medium-Sized Enterprises Phase 2 RAIL Grant No. 673780; EUROSTARS, DART Grant No. E10116 and DECIDE Grant No. E11541; the European Program PREDICT ITN Grant No. 766276; Third Joint Transnational Call 2016 JTC2016 ?CLEARLY? Grant No. UM 2017-8295; Interreg V-A Euregio Meuse-Rhine ?Euradiomics? Grant No. EMR4; and the Scientific Exchange from Swiss National Science Foundation Grant No. IZSEZ0_180524. We thank Simone Moorman for editing the manuscript.
Funding Information:
8295; Interreg V-A Euregio Meuse-Rhine “Euradiomics” Grant No. EMR4; and the Scientific Exchange from Swiss National Science Foundation Grant No. IZSEZ0_180524.
Publisher Copyright:
© 2020 by American Society of Clinical Oncology
PY - 2020/3/5
Y1 - 2020/3/5
N2 - Big data for health care is one of the potential solutions to deal with the numerous challenges of health care, such as rising cost, aging population, precision medicine, universal health coverage, and the increase of non-communicable diseases. However, data centralization for big data raises privacy and regulatory concerns.Covered topics include ( 1) an introduction to privacy of patient data and distributed learning as a potential solution to preserving these data, a description of the legal context for patient data research, and a definition of machine/deep learning concepts; (2) a presentation of the adopted review protocol; ( 3) a presentation of the search results; and (4) a discussion of the findings, limitations of the review, and future perspectives.Distributed learning from federated databases makes data centralization unnecessary. Distributed algorithms iteratively analyze separate databases, essentially sharing research questions and answers between databases instead of sharing the data. In other words, one can learn from separate and isolated datasets without patient data ever leaving the individual clinical institutes.Distributed learning promises great potential to facilitate big data for medical application, in particular for international consortiums. Our purpose is to review the major implementations of distributed learning in health care. (c) 2020 by American Society of Clinical Oncology
AB - Big data for health care is one of the potential solutions to deal with the numerous challenges of health care, such as rising cost, aging population, precision medicine, universal health coverage, and the increase of non-communicable diseases. However, data centralization for big data raises privacy and regulatory concerns.Covered topics include ( 1) an introduction to privacy of patient data and distributed learning as a potential solution to preserving these data, a description of the legal context for patient data research, and a definition of machine/deep learning concepts; (2) a presentation of the adopted review protocol; ( 3) a presentation of the search results; and (4) a discussion of the findings, limitations of the review, and future perspectives.Distributed learning from federated databases makes data centralization unnecessary. Distributed algorithms iteratively analyze separate databases, essentially sharing research questions and answers between databases instead of sharing the data. In other words, one can learn from separate and isolated datasets without patient data ever leaving the individual clinical institutes.Distributed learning promises great potential to facilitate big data for medical application, in particular for international consortiums. Our purpose is to review the major implementations of distributed learning in health care. (c) 2020 by American Society of Clinical Oncology
KW - ONCOLOGY
KW - STATISTICS
U2 - 10.1200/CCI.19.00047
DO - 10.1200/CCI.19.00047
M3 - (Systematic) Review article
C2 - 32134684
SN - 2473-4276
VL - 4
SP - 184
EP - 200
JO - JCO Clinical Cancer Informatics
JF - JCO Clinical Cancer Informatics
ER -