SPINEPS-automatic whole spine segmentation of T2-weighted MR images using a two-phase approach to multi-class semantic and instance segmentation.

Journal: European radiology
PMID:

Abstract

OBJECTIVES: Introducing SPINEPS, a deep learning method for semantic and instance segmentation of 14 spinal structures (ten vertebra substructures, intervertebral discs, spinal cord, spinal canal, and sacrum) in whole-body sagittal T2-weighted turbo spin echo images.

Authors

  • Hendrik Möller
    Institute for Neuroradiology, TUM University Hospital, School of Medicine and Health, Technical University of Munich (TUM), Munich, Germany; Chair for AI in Healthcare and Medicine, Technical University of Munich (TUM) and TUM University Hospital, Munich, Germany.
  • Robert Graf
    Institute for Neuroradiology, TUM University Hospital, School of Medicine and Health, Technical University of Munich (TUM), Munich, Germany; Chair for AI in Healthcare and Medicine, Technical University of Munich (TUM) and TUM University Hospital, Munich, Germany.
  • Joachim Schmitt
    Department of Diagnostic and Interventional Neuroradiology, School of Medicine and Health, TUM Klinikum Rechts der Isar, Technical University of Munich, Munich, Germany.
  • Benjamin Keinert
    Department of Diagnostic and Interventional Neuroradiology, School of Medicine and Health, Technical University of Munich, Munich, Germany.
  • Hanna Schön
    Department of Diagnostic and Interventional Neuroradiology, School of Medicine and Health, Technical University of Munich, Munich, Germany.
  • Matan Atad
    Institute for Neuroradiology, TUM University Hospital, School of Medicine and Health, Technical University of Munich (TUM), Munich, Germany; Chair for AI in Healthcare and Medicine, Technical University of Munich (TUM) and TUM University Hospital, Munich, Germany. Electronic address: matan.atad@tum.de.
  • Anjany Sekuboyina
  • Felix Streckenbach
    Institut für Diagnostische und Interventionelle Radiologie, Kinder- und Neuroradiologie, Universitätsmedizin Rostock, Deutschland.
  • Florian Kofler
    Department of Computer Science, Institute for AI in Medicine, Technical University of Munich, Munich, Germany; Department of Diagnostic and Interventional Neuroradiology, School of Medicine, Technical University of Munich, Munich, Germany; TranslaTUM, Central Institute for Translational Cancer Research of the Technical University of Munich, Munich, Germany; Helmholtz AI, Helmholtz Munich, Neuherberg, Germany.
  • Thomas Kroencke
    Department of Diagnostic and Interventional Radiology, University Hospital Augsburg, Augsburg, Germany.
  • Stefanie Bette
    Department of Diagnostic and Interventional Radiology, University Hospital Augsburg, Augsburg, Germany.
  • Stefan N Willich
    Institute of Social Medicine, Epidemiology and Health Economics, Charité-Universitätsmedizin Berlin, Berlin, Germany.
  • Thomas Keil
    Institute of Social Medicine, Epidemiology and Health Economics, Charité-Universitätsmedizin Berlin, Berlin, Germany.
  • Thoralf Niendorf
    Berlin Ultrahigh Field Facility, Max Delbrück Center for Molecular Medicine in the Helmholtz Association, Berlin, Germany.
  • Tobias Pischon
    Max-Delbrueck-Center for Molecular Medicine in the Helmholtz Association (MDC), Molecular Epidemiology Research Group, Berlin, Germany.
  • Beate Endemann
    Max Delbrück Center for Molecular Medicine in the Helmholtz Association, Berlin, Germany.
  • Bjoern Menze
  • Daniel Rueckert
    Biomedical Image Analysis (BioMedIA) Group, Department of Computing, Imperial College London, UK. Electronic address: d.rueckert@imperial.ac.uk.
  • Jan S Kirschke
    Neuroradiology, School of Medicine, Technical University of Munich, Munich, Germany.