Glossary

Federated learning in healthcare

What federated learning means when the data is patient data — how institutions train a shared model without moving records, what it has already delivered across dozens of hospitals, and why it is a privacy tool rather than a privacy guarantee. As of September 2026.

By Jonas WeirReviewed by Jonas Weir · editorial reviewUpdated

The short version

  • Federated learning trains one shared model across many institutions while each keeps its own patient data behind its own firewall — only model updates travel, never records.
  • A multi-institutional healthcare study reached up to 99% of the accuracy of a model trained on centrally pooled data, without pooling the data.
  • The EXAM model, trained across 20 institutes, predicted COVID-19 oxygen needs at an average AUC above 0.92 and improved generalizability by 38% over single-site training.
  • Sharing model updates is a privacy improvement but not a privacy guarantee — updates can leak information, so federated systems add secure aggregation and differential privacy.

Federated learning is a machine-learning method that trains one shared model across many institutions without moving their data. Each site trains locally on its own patients; only model updates — never records — travel to a coordinator, which combines them into a single consensus model. As of September 2026.

The term was coined in 2017 to describe a decentralized alternative to collecting everything in one place: leave the training data where it lives, learn locally, and aggregate the locally-computed updates into a shared model 1. In healthcare the same idea answers a specific problem — the most valuable training data sits in separate hospital systems that cannot legally or practically merge their records.

Why it matters in healthcare

Patient data is both the fuel for clinical AI and the thing that regulation, consent, and institutional trust most tightly restrict. A model trained on one hospital's population tends to generalize poorly to another's. Federated learning lets institutions gain insights collaboratively — in the form of a consensus model — without moving patient data beyond the firewalls of the sites where it resides 2. That reframes the classic trade-off: you no longer have to choose between a bigger, more representative training set and keeping data under local governance.

How it works in practice

A coordinator sends the current model to each participating site. Every site trains on its own data for a round, then returns only the updated model parameters. The coordinator averages those updates into a new shared model and repeats. Raw records never leave the building. Because sites differ in scanner hardware, coding habits, and patient mix, real deployments add steps to handle that heterogeneity, and they layer on privacy protections described below.

Where it appears today

The evidence base is now built on real multi-site systems, not demonstrations. A multi-institutional study showed federated models reaching up to 99% of the accuracy of a model trained on the same data pooled centrally — without any site sharing its data 3. During the pandemic, the EXAM model was trained across 20 institutes worldwide to predict the oxygen needs of patients with COVID-19, achieving an average AUC above 0.92 and a 38% average gain in generalizability over models trained at a single site 4. The largest reported effort assembled data from 71 sites across six continents — 6,314 glioblastoma cases — to train a rare-cancer boundary detector that no single institution had the volume to build alone 5.

Common misunderstandings

The most consequential error is treating federated learning as automatic privacy. Keeping raw records local removes the biggest exposure, but the model updates that do travel can still leak information about the training data. Serious systems therefore pair the federated setup with secure aggregation, so the coordinator sees only the combined update, and differential privacy, which adds calibrated noise to bound what any update reveals 6. A second misunderstanding is assuming a federated model equals a centralized one; site heterogeneity, uneven data quality, and coordination overhead mean results have to be validated, not assumed. Federated learning widens the training set — it does not retire the need for external validation.

Related terms

Federated learning sits alongside other techniques for training on sensitive records. Synthetic patient data generates artificial records that stand in for real ones, while de-identification versus anonymization governs when stripped data is still legally personal. Many federated projects now train or fine-tune a foundation model rather than a single-task network. For the broader numbers on where clinical AI has spread, see our AI in healthcare statistics.

We revisit this page on a yearly cycle, and sooner if a new multi-institutional deployment, a regulator's guidance on federated training, or a demonstrated attack on shared updates changes the picture.

Questions & answers

  • Does federated learning move patient data between hospitals?

    No. Each institution trains a copy of the model on its own patients behind its own firewall, and only the resulting model updates — the adjusted numbers, never the records — are sent to a coordinator that combines them. The data stays where its governance stays.

  • Is federated learning enough to protect patient privacy on its own?

    It is a strong improvement over pooling raw data, but it is a privacy tool rather than a privacy guarantee. Shared model updates can leak information about the underlying records, which is why serious deployments add secure aggregation and differential privacy on top of the federated setup.

  • How well does a federated model perform compared with centralized training?

    In a landmark multi-institutional study, models trained with federated learning reached up to 99% of the accuracy of a model trained on the same data pooled centrally — while the data never left each site.

Sources

  1. McMahan HB, Moore E, Ramage D, Hampson S, Aguera y Arcas B. Communication-Efficient Learning of Deep Networks from Decentralized Data. Proceedings of the 20th International Conference on Artificial Intelligence and Statistics (AISTATS). 2017;54:1273-1282. proceedings.mlr.press/v54/mcmahan17a.html
  2. Rieke N, Hancox J, Li W, et al. The future of digital health with federated learning. npj Digital Medicine. 2020;3:119. doi.org/10.1038/s41746-020-00323-1
  3. Sheller MJ, Edwards B, Reina GA, et al. Federated learning in medicine: facilitating multi-institutional collaborations without sharing patient data. Scientific Reports. 2020;10:12598. doi.org/10.1038/s41598-020-69250-1
  4. Dayan I, Roth HR, Zhong A, et al. Federated learning for predicting clinical outcomes in patients with COVID-19. Nature Medicine. 2021;27:1735-1743. doi.org/10.1038/s41591-021-01506-3
  5. Pati S, Baid U, Edwards B, et al. Federated learning enables big data for rare cancer boundary detection. Nature Communications. 2022;13:7346. doi.org/10.1038/s41467-022-33407-5
  6. Kaissis GA, Makowski MR, Ruckert D, Braren RF. Secure, privacy-preserving and federated machine learning in medical imaging. Nature Machine Intelligence. 2020;2:305-311. doi.org/10.1038/s42256-020-0186-1