Header

UZH-Logo

Maintenance Infos

Federated learning enables big data for rare cancer boundary detection


Pati, Sarthak; Baid, Ujjwal; Edwards, Brandon; et al; Li, Hongwei; Weiss, Tobias; Weller, Michael (2022). Federated learning enables big data for rare cancer boundary detection. Nature Communications, 13(1):7346.

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.

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.

Statistics

Citations

Dimensions.ai Metrics
60 citations in Web of Science®
54 citations in Scopus®
Google Scholar™

Altmetrics

Downloads

16 downloads since deposited on 01 Feb 2023
13 downloads since 12 months
Detailed statistics

Additional indexing

Item Type:Journal Article, refereed, original work
Communities & Collections:04 Faculty of Medicine > University Hospital Zurich > Clinic for Neurology
07 Faculty of Science > Department of Quantitative Biomedicine
04 Faculty of Medicine > University Hospital Zurich > Clinic for Neuroradiology
Dewey Decimal Classification:610 Medicine & health
Scopus Subject Areas:Physical Sciences > General Chemistry
Life Sciences > General Biochemistry, Genetics and Molecular Biology
Health Sciences > Multidisciplinary
Physical Sciences > General Physics and Astronomy
Language:English
Date:5 December 2022
Deposited On:01 Feb 2023 08:28
Last Modified:29 May 2024 01:46
Publisher:Nature Publishing Group
ISSN:2041-1723
OA Status:Gold
Free access at:Publisher DOI. An embargo period may apply.
Publisher DOI:https://doi.org/10.1038/s41467-022-33407-5
PubMed ID:36470898
Other Identification Number:Erratum in: 10.1038/s41467-023-36188-7 (DOI), 36702828 (PubMed ID)
  • Content: Published Version
  • Licence: Creative Commons: Attribution 4.0 International (CC BY 4.0)
  • Content: Published Version
  • Language: English
  • Description: Erratum
  • Licence: Creative Commons: Attribution 4.0 International (CC BY 4.0)