Boosting multi-demographic federated learning for chest radiograph analysis using general-purpose self-supervised representations

European Journal of Radiology Artificial Intelligence, vol. 3, pp. 100028

Abstract

Background Reliable artificial intelligence (AI) models for medical image analysis often depend on large and diverse labeled datasets. Federated learning (FL) offers a decentralized and privacy-preserving approach to training but struggles in highly non-independent and identically distributed (non-IID) settings, where institutions with more representative data may experience degraded performance. Moreover, existing large-scale FL studies have been limited to adult datasets, neglecting the unique challenges posed by pediatric data, which introduces additional non-IID variability. Methods To address these limitations, we analyzed n=398,523 adult chest radiographs from diverse institutions across multiple countries and n=9,125 pediatric images, leveraging transfer learning from general-purpose self-supervised image representations to classify pneumonia and cases with no abnormality. Results Using state-of-the-art vision transformers, we found that FL improved performance only for smaller adult datasets (P<0.001) but degraded performance for larger datasets (P≤0.063) and pediatric cases (P=0.242). However, equipping FL with self-supervised weights significantly enhanced outcomes across pediatric cases (P=0.031) and most adult datasets (P≤0.007), except the largest dataset (P=0.052). Conclusions These findings underscore the potential of easily deployable general-purpose self-supervised image representations to address non-IID challenges in clinical FL applications and highlight their promise for enhancing patient outcomes and advancing pediatric healthcare, where data scarcity and variability remain persistent obstacles.

Authors 4

1. Mahshad Lotfinia Aachen Department of Diagnostic and Interventional Radiology

   Universitätsklinikum Aachen · RWTH Aachen University

   Affiliation as printed

   Department of Diagnostic and Interventional Radiology, University Hospital RWTH Aachen, Aachen, Germany

2. Arash Tayebiarasteh

   Hamedan University of Technology

   Affiliation as printed

   Department of Computer Engineering, Hamedan University of Technology, Hamedan, Iran

3. Mehdi Joodaki Aachen Institute for Computational Genomics Joint Research Center for Computational Biomedicine

   Universitätsklinikum Aachen · RWTH Aachen University

   Affiliation as printed

   Institute for Computational Genomics, Joint Research Center for Computational Biomedicine, University Hospital RWTH Aachen, Aachen, Germany

4. Soroosh Tayebi Arasteh corresponding Aachen Department of Diagnostic and Interventional Radiology

   Friedrich-Alexander-Universität Erlangen-Nürnberg · Universitätsklinikum Aachen · Stanford Medicine · RWTH Aachen University · Stanford University

   Affiliation as printed

   Department of Diagnostic and Interventional Radiology, University Hospital RWTH Aachen, Aachen, Germany

   Department of Radiology, Stanford University, Stanford, CA, USA

   Department of Urology, Stanford University, Stanford, CA, USA

   Pattern Recognition Lab, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany