An international study presenting a federated learning AI platform for pediatric brain tumors

While multiple factors impact disease, artificial intelligence (AI) studies in medicine often use small, non-diverse patient cohorts due to data sharing and privacy issues. Federated learning (FL) has emerged as a solution, enabling training across hospitals without direct data sharing. Here, we present FL-PedBrain, an FL platform for pediatric posterior fossa brain tumors, and evaluate its performance on a diverse, realistic, multi-center cohort. Pediatric brain tumors were targeted due to the scarcity of such datasets, even in tertiary care hospitals. Our platform orchestrates federated training for joint tumor classification and segmentation across 19 international sites. FL-PedBrain exhibits less than a 1.5% decrease in classification and a 3% reduction in segmentation performance compared to centralized data training. FL boosts segmentation performance by 20 to 30% on three external, out-of-network sites. Finally, we explore the sources of data heterogeneity and examine FL robustness in real-world scenarios with data imbalances.

Authors

Institutions

Publication Details

Journal
Nature Communications
Published
2024-09-02
DOI
https://doi.org/10.1038/s41467-024-51172-5
Citations
33
Primary Topic
Artificial Intelligence in Healthcare and Education
Type
article
Field-Weighted Citation Impact
3.97

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

An international study presenting a federated learning AI platform for pediatric brain tumors

Edward H. Lee, Tina Young Poussaint, Maryam Mohammadzadeh, Kristen W. Yeom et al.
33 citations
Nature Communications
Artificial Intelligence in Healthcare and Education
3.97
article

An international study presenting a federated learning AI platform for pediatric brain tumors

Edward H. Lee, Tina Young Poussaint, Maryam Mohammadzadeh, Kristen W. Yeom, Olivia Choudhury, Michelle Han, Gerald A. Grant, Arastoo Vossough, Michael Kuwabara, Cynthia Campen, Birgit Ertl‐Wagner, Laura M. Prolo, Hakan Doğan, Jason L. Wright, Alireza Radmanesh, Robert Paul Goetti, Nicholas A. Vitanza, Robert M. Lober, Gang Fu, Vijay Ramaswamy, Nils D. Forkert, Michael E. Moseley, John Curran, Samuel Cheshier, Azam Eghbal, Ujjwal Ratan, Özgür Öztekin, Matthias Wagner, Alok Jaju, S.S. Wong, Kshitij Mankad, Hollie Lai, Kristian Aquilina, Emre Altınmakas, Jeffrey H Miller, Mourad Ben Saïd, Kathryn A. Szymanski, Eric M. Thompson, Sebastian Toescu, Chang Yueh Ho, Sébastien Perreault, Michael Zhang, Jacob Mevorach, Mattias Wilms
article en
33 citations

Abstract

While multiple factors impact disease, artificial intelligence (AI) studies in medicine often use small, non-diverse patient cohorts due to data sharing and privacy issues. Federated learning (FL) has emerged as a solution, enabling training across hospitals without direct data sharing. Here, we present FL-PedBrain, an FL platform for pediatric posterior fossa brain tumors, and evaluate its performance on a diverse, realistic, multi-center cohort. Pediatric brain tumors were targeted due to the scarcity of such datasets, even in tertiary care hospitals. Our platform orchestrates federated training for joint tumor classification and segmentation across 19 international sites. FL-PedBrain exhibits less than a 1.5% decrease in classification and a 3% reduction in segmentation performance compared to centralized data training. FL boosts segmentation performance by 20 to 30% on three external, out-of-network sites. Finally, we explore the sources of data heterogeneity and examine FL robustness in real-world scenarios with data imbalances.

Nature CommunicationsVol. 15(1)
Creighton University (US), Boston Children's Hospital (US), Seattle Children's Hospital (US), Children's Hospital of Orange County (US), Amazon (United States) (US), Lucile Packard Children's Hospital (US), Children's Hospital of Philadelphia (US), Koç University (TR), University of Calgary (CA), University of Utah (US), Great Ormond Street Hospital (GB), Hospital for Sick Children (CA), University Hospital Augsburg (DE), Alberta Children's Hospital (CA), Centre Hospitalier Universitaire Sainte-Justine (CA), Children's Hospital at Westmead (AU), Riley Hospital for Children (US), Phoenix (United States) (US), Dayton Children's Hospital (US), Phoenix Children's Hospital (US), Kaiser Permanente West Los Angeles Medical Center (US), Duke Children's Hospital & Health Center (US), Izmir Tepecik Eğitim ve Araştırma Hastanesi (TR), Stanford Medicine (US), Kaiser Foundation Hospital (US), Kaiser Foundation Hospital (US), Hotchkiss Brain Institute (CA), University College London (GB), Hamad Medical Corporation (QA), New York University (US), Tehran University of Medical Sciences (IR), Université de Montréal (CA), Stanford University (US), Icahn School of Medicine at Mount Sinai (US)
Amazon Web Services, National Institute for Health and Care Research, National Institutes of Health, Canadian Institutes of Health Research, Great Ormond Street Institute of Child Health, Garron Family Cancer Centre, NIHR Great Ormond Street Hospital Biomedical Research Centre
Openalex Percentile: Top 6%
Artificial Intelligence in Healthcare and Education
3.97
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.