CoChat

The future of digital health with federated learning

Open access
Rieke, Nicola, Hancox, Jonny, Li, Wenqi, Milletarì, Fausto, Roth, Holger R., Albarqouni, Shadi, Bakas, Spyridon, Galtier, Mathieu N., Landman, Bennett A., Maier-Hein, Klaus, Ourselin, Sébastien, Sheller, Micah, Summers, Ronald M., Trask, Andrew, Xu, Daguang, Baust, Maximilian, Cardoso, M. Jorge · 2020 · npj Digital Medicine

The DOI matches its Crossref record; 74 of the 74 references it cites were found in a registry or an index; 30 of the 51 sources its claims cite were checked for support.

Abstract

Data-driven machine learning (ML) has emerged as a promising approach for building accurate and robust statistical models from medical data, which is collected in huge volumes by modern healthcare systems. Existing medical data is not fully exploited by ML primarily because it sits in data silos and privacy concerns restrict access to this data. However, without access to sufficient data, ML will be prevented from reaching its full potential and, ultimately, from making the transition from research to clinical practice. This paper considers key factors contributing to this issue, explores how federated learning (FL) may provide a solution for the future of digital health and highlights the challenges and considerations that need to be addressed.

Overview

CoChat Summary

TL;DR

This perspective paper argues that federated learning (FL) is a promising way to train medical AI across institutions without centralizing sensitive data, potentially improving accuracy, robustness, and privacy in digital health. It also emphasizes that FL does not remove core challenges such as data heterogeneity, privacy leakage, and the need for standardization and governance.

Key findings

  • FL is presented as a learning paradigm that lets multiple parties train collaboratively without exchanging or centralizing datasets, by moving the model to the data rather than the data to the model.
  • The paper states that recent research has shown FL models can achieve performance comparable to centrally hosted data sets and superior to models trained on isolated single-institution data.
  • Reported application areas already include EHR tasks such as finding clinically similar patients and predicting hospitalizations due to cardiac events, mortality, and ICU stay time, as well as medical imaging tasks such as whole-brain segmentation, brain tumor segmentation, and fMRI classification.
  • Concrete initiatives mentioned include a collaboration across 4 hospitals in France (HealthChain), an international federation of 30 healthcare institutions (FeTS), and a project across 10 pharmaceutical companies (Melloddy).

Methods

  • Not stated in the paper.
  • The article is a perspective/consensus paper rather than an empirical study; it synthesizes prior work and examples to discuss FL’s promise, current uses, and technical considerations.
  • The text describes FL workflows, including aggregation-server and peer-to-peer topologies, and discusses mechanisms such as FedAvg, FedProx, domain adaptation, and differential privacy at a conceptual level.

Limitations

  • The paper explicitly states that FL does not solve all issues inherent to learning on medical data; success still depends on data quality, bias, and standardization.
  • It states that FL does not eliminate privacy risks: models can still memorize information, and gradients/model updates may leak private data through inversion or adversarial attacks.
  • Evident from the text: this is a perspective article, so it does not report original experimental results, sample sizes, or statistical testing.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1038/s41746-020-00323-1

Cites

74 of 74 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Fig. 1 Example federated learning (FL) workflows and difference to learning on a Centralised Data Lake. a FL aggregation server—the typical FL workflow in which a federation of training nodes receive the global model, resubmit their partially trained models to a central server intermittently for aggregation and then continue training on the consensus model that the server returns. b FL peer to peer—alternative formulation of FL in which each training node exchanges its partially trained models with some or all of its peers and each does its own aggregation. c Centralised training—the general non-FL training workflow in which data acquiring sites donate their data to a central Data Lake from which they and others are able to extract data for local, independent training.
Fig. 1 Example federated learning (FL) workflows and difference to learning on a Centralised Data Lake. a FL aggregation server—the typical FL workflow in which a federation of training nodes receive the global model, resubmit their partially trained models to a central server intermittently for aggregation and then continue training on the consensus model that the server returns. b FL peer to peer—alternative formulation of FL in which each training node exchanges its partially trained models with some or all of its peers and each does its own aggregation. c Centralised training—the general non-FL training workflow in which data acquiring sites donate their data to a central Data Lake from which they and others are able to extract data for local, independent training.CC BY · PubMed Central
Fig. 2 Overview of different FL design choices. FL topologies—communication architecture of a federation. a Centralised: the aggregation server coordinates the training iterations and collects, aggregates and distributes the models to and from the Training Nodes (Hub & Spoke). b Decentralised: each training node is connected to one or more peers and aggregation occurs on each node in parallel. c Hierarchical: federated networks can be composed from several sub-federations, which can be built from a mix of Peer to Peer and Aggregation Server federations (d)). FL compute plans—trajectory of a model across several partners. e Sequential training/cyclic transfer learning. f Aggregation server, g Peer to Peer.
Fig. 2 Overview of different FL design choices. FL topologies—communication architecture of a federation. a Centralised: the aggregation server coordinates the training iterations and collects, aggregates and distributes the models to and from the Training Nodes (Hub & Spoke). b Decentralised: each training node is connected to one or more peers and aggregation occurs on each node in parallel. c Hierarchical: federated networks can be composed from several sub-federations, which can be built from a mix of Peer to Peer and Aggregation Server federations (d)). FL compute plans—trajectory of a model across several partners. e Sequential training/cyclic transfer learning. f Aggregation server, g Peer to Peer.CC BY · PubMed Central

Verification

Claims the paper makes

16 Valid · 0 Not valid · 4 Not checked · 19 Own results and interpretations

10 references still to check

A free account verifies the references that are still open.

Verify all 10 references — free account
  • Probably not a substantial claim. 19
  • It may state two separate claims as one. 32
  • The quote only partly states this claim. 1

Create a free account to see these claims.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

27 more claims about this paper

Create a free account to see every claim other papers make about this one.

See all claims — free account

Citations

Cites

Show all 74 references

Cited by

See all 4,015

Metadata

Authors
Rieke, N., Hancox, J., Li, W., Milletarì, F., Roth, H. R., Albarqouni, S., Bakas, S., Galtier, M. N., Landman, B. A., Maier-Hein, K., Ourselin, S., Sheller, M., Summers, R. M., Trask, A., Xu, D., Baust, M., & Cardoso, M. J.
Affiliations
  • Technical University of Munich (DE)
  • Nvidia (United Kingdom) (GB)
  • Nvidia (United States) (US)
  • Imperial College London (GB)
  • University of Pennsylvania (US)
  • Vanderbilt University (US)
  • German Cancer Research Center (DE)
  • Intel (United States) (US)
  • National Institutes of Health (US)
  • University of Oxford (GB)
  • King's College London (GB)
Corresponding author
Rieke, Nicola
Venue
npj Digital Medicine
Published
2020-09-14
Volume
3
Issue
1
Pages
119-119
ISSN
2398-6352 (ISSN-L)
OpenAlex
W3012501605
Cited by
4,015
References
38
Citations by year
2026: 1453 · 2025: 921 · 2024: 618 · 2023: 473 · 2022: 353 · 2021: 175 · 2020: 18
FWCI
13.36
Type
journal article
Language
en
Open access
Licence
Locations
  • npj Digital Medicine (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • arXiv (Cornell University) (submittedVersion, other-oa)
  • Informationszentrum Lebenswissenschaften (ZB MED) (submittedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • Research Portal (King's College London) (submittedVersion, other-oa)
  • mediaTUM – the media and publications repository of the Technical University Munich (Technical University Munich) (submittedVersion, other-oa)
  • arXiv (Cornell University)
Funders
  • U.S. Department of Health and Human Services (R01NS042645, U01CA242871)
  • Nvidia
  • Wellcome Trust
  • UK Research and Innovation
  • Bpifrance
  • Deutscher Akademischer Austauschdienst
  • Bundesministerium für Bildung und Forschung
  • National Institutes of Health (1s10od023495-01)
  • Engineering and Physical Sciences Research Council ([WT203148/Z/16/Z], WT213038/Z/18/Z)
  • Centre For Medical Engineering, King’s College London (WT 203148/Z/16/Z)
  • National Cancer Institute (U24CA189523, U01CA242871)
  • National Institute of Neurological Disorders and Stroke (R01NS042645)
  • NIH Clinical Center
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
  • Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
  • Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
Keywords
federated learning, digital health, medical data privacy, data silos, data-driven machine learning, health information access

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: The future of digital health with federated learning. https://app.cochat.ai/papers/10.1038/s41746-020-00323-1
The paper
Rieke, N., Hancox, J., Li, W., Milletarì, F., Roth, H. R., Albarqouni, S., Bakas, S., Galtier, M. N., Landman, B. A., Maier-Hein, K., Ourselin, S., Sheller, M., Summers, R. M., Trask, A., Xu, D., Baust, M., & Cardoso, M. J. (2020). The future of digital health with federated learning. npj Digital Medicine. https://doi.org/10.1038/s41746-020-00323-1

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.