https://doi.org/10.1140/epjs/s11734-026-02475-9
Review
Federated learning in healthcare: a review of the deployment gap between simulated and real-world implementations
1
Research Institute of Applied Artificial Intelligence and Digital Solutions, Plekhanov Russian University of Economics, Stremyanny per., 36, 115054, Moscow, Russia
2
Engineering Academy, Peoples’ Friendship University of Russia named after Patrice Lumumba, Miklukho-Maklaya St., 6, 117198, Moscow, Russia
3
Advanced Engineering School “Intelligent Theranostic Systems”, Sechenov University, Trubetskaya St., 8s2, 119991, Moscow, Russia
a
This email address is being protected from spambots. You need JavaScript enabled to view it.
Received:
28
May
2026
Accepted:
19
June
2026
Published online:
28
June
2026
Abstract
Federated learning (FL) has emerged as a promising paradigm for privacy-preserving collaborative model training in healthcare, yet a pronounced gap persists between its performance in simulated research settings and genuine clinical deployments. This scoping review, conducted following PRISMA-ScR guidelines, systematically surveyed FL research in healthcare indexed in PubMed from 2016 onward. Of 1,338 initially identified papers, 772 met inclusion criteria, of which only 25 (3.2%) represented genuine real-world FL deployments. Real-world deployments generally demonstrated near-equivalent performance to centralized baselines, though practical barriers, such as infrastructure complexity, data heterogeneity, partial labeling, and unvalidated privacy mechanisms, remain pervasive. These findings underscore that the federated learning deployment gap is primarily an infrastructural and organizational challenge rather than an algorithmic one, and identify key conditions required for broader clinical adoption.
Copyright comment Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
Elena Pitsik, Ivan Shishkin, Alsu Ivanova, Nikita Muchanko, Ilya Khanchuk, Fu Xiren, Denis Andrikov and Alexander Hramov have contributed equally to this work.
© The Author(s), under exclusive licence to EDP Sciences, Springer-Verlag GmbH Germany, part of Springer Nature 2026
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
