A O H Gerstner1, W Laffers, G Schade, F Göke, R Martin, B Thies. 1. Klinik und Poliklinik für Hals-Nasen-Ohren-Heilkunde/Chirurgie, Universität Bonn, Sigmund-Freud-Str. 25, 53105, Bonn, Deutschland. andreas.gerstner@ukb.uni-bonn.de
Abstract
BACKGROUND: Hyperspectral imaging has been proven to be useful in remote earth sensing, e.g., satellite-based classification of vegetation. After modifying it for in vivo evaluation of the larynx using microscopy, we show its adoption for endoscopes. MATERIALS AND METHODS: In addition to routine microlaryngoscopies under white light inspection, the laryngoscopy was also performed with not only a conventional operation microscope mounted with a tuneable monochromatic light source and a synchronously triggered monochromatic CCD camera (n = 47 patients), but also a similarly modified rigid 0° endoscope (n = 16 patients). Hyperspectral image cubes were obtained between 390 and 680 nm, analyzed using established software tools, and the data using microscope versus endoscope were compared. RESULTS: Under endoscopy, illumination was more even and sterical resolution appeared unchanged at significantly shorter image acquisition times. Hyperspectral classification delineated areas of altered mucosa without further external information irrespective of whether image cubes were generated with microscopy or endoscopy. CONCLUSION: Hyperspectral imaging can be transferred not only to micro(laryngo)scopy but also to endoscopy with rigid optics. This opens the way to a variety of clinically relevant anatomical regions (e.g., upper aerodigestive tract). Its adoption to flexible optics will further broaden this spectrum.
BACKGROUND: Hyperspectral imaging has been proven to be useful in remote earth sensing, e.g., satellite-based classification of vegetation. After modifying it for in vivo evaluation of the larynx using microscopy, we show its adoption for endoscopes. MATERIALS AND METHODS: In addition to routine microlaryngoscopies under white light inspection, the laryngoscopy was also performed with not only a conventional operation microscope mounted with a tuneable monochromatic light source and a synchronously triggered monochromatic CCD camera (n = 47 patients), but also a similarly modified rigid 0° endoscope (n = 16 patients). Hyperspectral image cubes were obtained between 390 and 680 nm, analyzed using established software tools, and the data using microscope versus endoscope were compared. RESULTS: Under endoscopy, illumination was more even and sterical resolution appeared unchanged at significantly shorter image acquisition times. Hyperspectral classification delineated areas of altered mucosa without further external information irrespective of whether image cubes were generated with microscopy or endoscopy. CONCLUSION: Hyperspectral imaging can be transferred not only to micro(laryngo)scopy but also to endoscopy with rigid optics. This opens the way to a variety of clinically relevant anatomical regions (e.g., upper aerodigestive tract). Its adoption to flexible optics will further broaden this spectrum.
Authors: Jared L Schmidt; Jonathan C Henriksen; Dan M McKeon; Kay Savik; H Evin Gulbahce; Stefan E Pambuccian Journal: Cancer Date: 2008-10-25 Impact factor: 6.860
Authors: Omer Khalid; Sofyan Radaideh; Oscar W Cummings; Michael J O'Brien; John R Goldblum; Douglas K Rex Journal: World J Gastroenterol Date: 2009-08-14 Impact factor: 5.742
Authors: J Löhler; A O H Gerstner; F Bootz; H Heinritz; A Fryen; G Fryen; N Holstein; A Lingg; J Kleeberg; W Langhoff; G Rösch; A Hanisch; E Schneeberg; D Heinrich; L E Walther Journal: HNO Date: 2012-03 Impact factor: 1.284
Authors: Wiebke Laffers; Ann-Christin Busse; Jens Mahrt; Phuc Nguyen; Andreas O H Gerstner; Friedrich Bootz; Johannes T Wessels Journal: Eur Arch Otorhinolaryngol Date: 2014-04-01 Impact factor: 2.503