Sparse keypoint segmentation of lung fissures: efficient geometric deep learning for abstracting volumetric images.

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

Abstract

PURPOSE: Lung fissure segmentation on CT images often relies on 3D convolutional neural networks (CNNs). However, 3D-CNNs are inefficient for detecting thin structures like the fissures, which make up a tiny fraction of the entire image volume. We propose to make lung fissure segmentation more efficient by using geometric deep learning (GDL) on sparse point clouds.

Authors

  • Paul Kaftan
    Institute of Medical Systems Biology, Ulm University, Albert-Einstein-Allee 11, 89081, Ulm, Germany.
  • Mattias P Heinrich
    Institute of Medical Informatics, University of Lübeck, Ratzeburger Allee 160, 23562, Lübeck, Germany. heinrich@imi.uni-luebeck.de.
  • Lasse Hansen
    Institute of Medical Informatics, University of Lübeck, Ratzeburger Allee 160, 23562, Lübeck, Germany. hansen@imi.uni-luebeck.de.
  • Volker Rasche
    Department of Internal Medicine II, Ulm University Medical Center, Albert-Einstein-Allee 23, 89081, Ulm, Germany.
  • Hans A Kestler
    Institute of Medical Systems Biology, Ulm University, Albert-Einstein-Allee 11, 89081 Ulm, Germany. hans.kestler@uni-ulm.de.
  • Alexander Bigalke
    Institute of Medical Informatics, University of Lübeck, Ratzeburger Allee 160, 23538, Lübeck, Germany. alexander.bigalke@uni-luebeck.de.