Tissue classification for laparoscopic image understanding based on multispectral texture analysis

• Verfasst von: Zhang, Yan [VerfasserIn]
• Verfasst von: Kenngott, Hannes Götz [VerfasserIn]
• Verfasst von: Wagner, Martin [VerfasserIn]
• Verfasst von: Mayer, Benjamin [VerfasserIn]
• Verfasst von: Stock, Christian [VerfasserIn]
• Verfasst von: Maier-Hein, Lena [VerfasserIn]
• Titel: Tissue classification for laparoscopic image understanding based on multispectral texture analysis
• Verf.angabe: Yan Zhang, Sebastian Wirkert, Justin Iszatt, Hannes Kenngott, Martin Wagner, Benjamin Mayer, Christian Stock, Neil T. Clancy, Daniel S. Elson, Lena Maier-Hein
• E-Jahr: 2017
• Jahr: 25 January 2017
• Umfang: 10 S.
• Fussnoten: Gesehen am 01.08.2018
• Titel Quelle: Enthalten in: Journal of medical imaging
• Ort Quelle: [Bellingham, Wash.] : SPIE, 2014
• Jahr Quelle: 2017
• Band/Heft Quelle: 4(2017,1) Artikel-Nummer 015001, 10 Seiten
• ISSN Quelle: 2329-4310
• DOI: doi:10.1117/1.JMI.4.1.015001
• Datenträger: Online-Ressource
• Sprache: eng
• K10plus-PPN: 1578151848

Abstract

Intraoperative 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 <italic>ex vivo</italic> study through statistical analysis, we show that (1) multispectral imaging data are superior to RGB data for organ tissue classification when used in conjunction with widely applied feature descriptors and (2) combining the tissue texture with the reflectance spectrum improves the classification performance. The classifier reaches an accuracy of 98.4% on our dataset. Multispectral tissue analysis could thus evolve as a key enabling technique in computer-assisted laparoscopy.