Tissue classification for laparoscopic image understanding based on multispectral texture analysis

Research output: Contribution to book/Conference proceedings/Anthology/ReportConference contributionContributedpeer-review

Contributors

  • Yan Zhang - , German Cancer Research Center (DKFZ) (Author)
  • Sebastian J. Wirkert - , German Cancer Research Center (DKFZ) (Author)
  • Justin Iszatt - , German Cancer Research Center (DKFZ) (Author)
  • Hannes Kenngott - , Heidelberg University  (Author)
  • Martin Wagner - , University Hospital Heidelberg (Author)
  • Benjamin Mayer - , Heidelberg University  (Author)
  • Christian Stock - , Heidelberg University  (Author)
  • Neil T. Clancy - , Imperial College London (Author)
  • Daniel S. Elson - , Imperial College London (Author)
  • Lena Maier-Hein - , German Cancer Research Center (DKFZ) (Author)

Abstract

Intra-operative tissue classification is one of the prerequisites for providing context-aware visualization in computer-assisted minimally invasive surgeries. As many anatomical structures are difficult to differentiate in conventional RGB medical images, we propose a classification method based on multispectral image patches. In a comprehensive ex vivo study we show (1) that multispectral imaging data is superior to RGB data for organ tissue classification when used in conjunction with widely applied feature descriptors and (2) that combining the tissue texture with the reflectance spectrum improves the classification performance. Multispectral tissue analysis could thus evolve as a key enabling technique in computer-assisted laparoscopy.

Details

Original languageEnglish
Title of host publicationMedical Imaging 2016
EditorsRobert J. Webster, Ziv R. Yaniv
PublisherSPIE - The international society for optics and photonics, Bellingham
ISBN (electronic)9781510600218
Publication statusPublished - 2016
Peer-reviewedYes
Externally publishedYes

Publication series

SeriesProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume9786
ISSN1605-7422

Conference

TitleMedical Imaging 2016: Image-Guided Procedures, Robotic Interventions, and Modeling
Duration28 February - 1 March 2016
CitySan Diego
CountryUnited States of America

Keywords

Keywords

  • multispectral laparoscopy, multispectral texture analysis, tissue classification