A transfer learning approach to facilitate ComBat-based harmonization of multicentre radiomic features in new datasets.

Journal: PloS one
PMID: 34197503

Abstract

PURPOSE: To facilitate the demonstration of the prognostic value of radiomics, multicenter radiomics studies are needed. Pooling radiomic features of such data in a statistical analysis is however challenging, as they are sensitive to the variability in scanner models, acquisition protocols and reconstruction settings, which is often unavoidable in a multicentre retrospective analysis. A statistical harmonization strategy called ComBat was utilized in radiomics studies to deal with the "center-effect". The goal of the present work was to integrate a transfer learning (TL) technique within ComBat-and recently developed alternate versions of ComBat with improved flexibility (M-ComBat) and robustness (B-ComBat)-to allow the use of a previously determined harmonization transform to the radiomic feature values of new patients from an already known center.

Authors

  • Ronrick Da-Ano
    INSERM, UMR 1101, LaTIM, University of Brest, Brest, France.
  • Francois Lucia
    LaTIM, INSERM, UMR 1101, University of Brest, CHRU Brest, France.
  • Ingrid Masson
    INSERM, UMR 1101, LaTIM, University of Brest, Brest, France.
  • Ronan Abgral
    Department of Nuclear Medicine, University of Brest, Brest, France.
  • Joanne Alfieri
    Department of Radiation Oncology, McGill University Health Centre, Montreal, Quebec.
  • Caroline Rousseau
  • Augustin Mervoyer
    Department of Radiation Oncology, Institut de cancérologie de l'Ouest René-Gauducheau, Saint-Herblain, France.
  • Caroline Reinhold
    Department of Radiology, McGill University Health Center, Montréal, Québec, Canada.
  • Olivier Pradier
    LaTIM, INSERM, UMR 1101, Brest 29609, France and Radiotherapy Department, CHRU Morvan, Brest 29609, France.
  • Ulrike Schick
    The Royal Marsden NHS Foundation Trust, London, UK.
  • Dimitris Visvikis
    LaTIM, INSERM, UMR 1101, Brest 29609, France.
  • Mathieu Hatt
    LaTIM, INSERM, UMR 1101, Brest 29609, France.

Keywords

External Resources

Popular Topics
Recent Journals