DenseSeg: joint learning for semantic segmentation and landmark detection using dense image-to-shape representation.

Journal: International journal of computer assisted radiology and surgery
PMID: 39849288

Abstract

PURPOSE: Semantic segmentation and landmark detection are fundamental tasks of medical image processing, facilitating further analysis of anatomical objects. Although deep learning-based pixel-wise classification has set a new-state-of-the-art for segmentation, it falls short in landmark detection, a strength of shape-based approaches.

Authors

  • Ron Keuth
    Medical Informatics, University of Lübeck, Ratzeburger Allee 160, 23562, Lübeck, Germany.
  • Lasse Hansen
    Institute of Medical Informatics, University of Lübeck, Ratzeburger Allee 160, 23562, Lübeck, Germany. hansen@imi.uni-luebeck.de.
  • Maren Balks
    Paediatric Surgery, University Hospital Schleswig-Holstein, Ratzeburger Allee 160, 23562, Lübeck, Germany.
  • Ronja Jäger
    Paediatric Surgery, University Hospital Schleswig-Holstein, Ratzeburger Allee 160, 23562, Lübeck, Germany.
  • Anne-Nele Schröder
    Paediatric Surgery, University Hospital Schleswig-Holstein, Ratzeburger Allee 160, 23562, Lübeck, Germany.
  • Ludger Tüshaus
    Paediatric Surgery, University Hospital Schleswig-Holstein, Ratzeburger Allee 160, 23562, Lübeck, Germany.
  • Mattias Heinrich

Keywords

External Resources

Popular Topics
Recent Journals