Multi-institutional PET/CT image segmentation using federated deep transformer learning

Computer Methods and Programs in Biomedicine, vol. 240, pp. 107706

Abstract

Generalizable and trustworthy deep learning models for PET/CT image segmentation require large heterogeneous multi-institutional datasets. However, legal, ethical, and patient privacy issues challenge sharing of datasets between centers. To overcome these challenges, we developed a federated learning (FL) framework for multi-institutional PET/CT image segmentation. A dataset consisting of 328 FL (HN) cancer patients who underwent clinical PET/CT examinations gathered from six different centers was enrolled. A pure transformer network was implemented as fully core segmentation algorithms using dual channel PET/CT images. We evaluated different frameworks (single center-based, centralized baseline, as well as seven different FL algorithms) using 68 PET/CT images (20% of each center data). In particular, the implemented FL algorithms include clipping with the quantile estimator (ClQu), zeroing with the quantile estimator (ZeQu), federated averaging (FedAvg), lossy compression (LoCo), robust aggregation (RoAg), secure aggregation (SeAg), and Gaussian differentially private FedAvg with adaptive quantile clipping (GDP-AQuCl). The Dice coefficient was 0.80±0.11 for both centralized and SeAg FL algorithms. All FL approaches achieved centralized learning model performance with no statistically significant differences. Among the FL algorithms, SeAg and GDP-AQuCl performed better than the other techniques. However, there was no statistically significant difference. All algorithms, except the center-based approach, resulted in relative errors less than 5% for SUVmax and SUVmean for all FL and centralized methods. Centralized and FL algorithms significantly outperformed the single center-based baseline. The developed FL-based (with centralized method performance) algorithms exhibited promising performance for HN tumor segmentation from PET/CT images.

Authors 10

1. Isaac Shiri

   University Hospital of Geneva

   Affiliation as printed

   Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, Geneva, Switzerland

2. Behrooz Razeghi

   University of Geneva

   Affiliation as printed

   Department of Computer Science, University of Geneva, Geneva, Switzerland

   Department of Computer Science, University of Geneva, Switzerland

3. Alireza Vafaei Sadr Aachen Institute of Pathology

   Pennsylvania State University · Universitätsklinikum Aachen · RWTH Aachen University

   Affiliation as printed

   Institute of Pathology, RWTH Aachen University Hospital, Aachen, Germany; Department of Public Health Sciences, College of Medicine, The Pennsylvania State University, Hershey, PA 17033, USA

   Institute of Pathology, RWTH Aachen University Hospital, Aachen, Germany

4. Mehdi Amini

   University Hospital of Geneva

   Affiliation as printed

   Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, Geneva, Switzerland

5. Yazdan Salimi

   University Hospital of Geneva

   Affiliation as printed

   Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, Geneva, Switzerland

6. Sohrab Ferdowsi

   University of Geneva

   Affiliation as printed

   Department of Computer Science, University of Geneva, Geneva, Switzerland

   Department of Computer Science, University of Geneva, Switzerland

7. Peter Boor Aachen Institute of Pathology

   Universitätsklinikum Aachen · RWTH Aachen University

   Affiliation as printed

   Institute of Pathology, RWTH Aachen University Hospital, Aachen, Germany

8. Denız Gündüz

   Imperial College London

   Affiliation as printed

   Department of Electrical and Electronic Engineering, Imperial College London, UK

9. Slava Voloshynovskiy

   University of Geneva

   Affiliation as printed

   Department of Computer Science, University of Geneva, Geneva, Switzerland

   Department of Computer Science, University of Geneva, Switzerland

10. Habib Zaidi corresponding

    University of Geneva · University of Groningen · University of Southern Denmark · University Hospital of Geneva

    Affiliation as printed

    Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, Geneva, Switzerland; Geneva University Neurocenter, University of Geneva, Geneva, Switzerland; Department of Nuclear Medicine and Molecular Imaging, University of Groningen, Groningen, The Netherlands; Department of Nuclear Medicine, University of Southern Denmark, Odense, Denmark. Electronic address: habib.zaidi@hcuge.ch

    Department of Nuclear Medicine and Molecular Imaging, University of Groningen, The Netherlands

    Department of Nuclear Medicine, University of Southern Denmark, Denmark

    Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, Geneva, Switzerland

    Geneva University Neurocenter, University of Geneva, Switzerland