CC BY-NC-ND 4.0 · Laryngorhinootologie 2019; 98(S 02): S255-S256
DOI: 10.1055/s-0039-1685961
Poster
Oncology

Hyperspectral imaging (HSI) for intraoperative tumor cell classification

R Beck
1   Klinik und Poliklinik für HNO-Heilkunde Universität Leipzig, Leipzig
,
H Köhler
2   Innovation Center Computer Assisted Surgery (ICCAS), Leipzig, Leipzig
,
F Goltz
3   Klinik für Viszeral-, Transplantations-, Thorax- und Gefäßchirurgie, Universitätsklinikum Leipzig AöR, Leipzig, Leipzig
,
A Dietz
4   Klinik und Poliklinik für Hals-, Nasen-, Ohrenheilkunde, Universitätsklinikum Leipzig AöR, Leipzig, Leipzig
,
C Chalopin
2   Innovation Center Computer Assisted Surgery (ICCAS), Leipzig, Leipzig
,
I Gockel
3   Klinik für Viszeral-, Transplantations-, Thorax- und Gefäßchirurgie, Universitätsklinikum Leipzig AöR, Leipzig, Leipzig
› Author Affiliations
 

Introduction:

Hyperspectral imaging (HSI) is an in vivo imaging technique that illuminates tissue with light in the visual and near-infrared spectrum and measures the light remitted by the tissue. Initial studies have shown promising results in evaluating tissue perfusion and tumor classification. We investigated the use of HSI for the differentiation of tumor tissue and tumor-free tissue.

methods:

Four squamous cell carcinomas of three patients were analyzed by HSI during laryngectomy. Examination of the preparations took place immediately after removal in the OR with the TIVITA ™ Tissue T2 camera system from Diaspective Vision GmbH (Am Salzhaff, Germany). Reflection spectra of tumor tissue and mucosa were recorded in the range of 500 – 1000nm. Standardization of the spectra with standard normal variate correction (SNV) and classification with a machine-learning method (SVM) were performed.

Results:

39,722 spectra from the tumor areas and 64,995 spectra from the mucosal areas were measured. The classification in mucous membrane and squamous cell carcinoma was possible with a correct classification rate of 0.88. Sensitivity and positive predictive value for squamous cell carcinomas were 0.78 and 0.88, respectively, and 0.94 and 0.88, respectively, for mucosal tissue.

Conclusion:

Squamous cell carcinomas have characteristic reflection spectra in the HSI. These could be used as a basis for a computer-assisted algorithm for endoscopic, non-invasive detection of squamous cell carcinoma. Validation requires further measurements in larger numbers. For the intraoperative application of HSI, the development of a system integrated into an endoscope would be advantageous.



Publication History

Publication Date:
23 April 2019 (online)

© 2019. The Author(s). This is an open access article published by Thieme under the terms of the Creative Commons Attribution-NonDerivative-NonCommercial-License, permitting copying and reproduction so long as the original work is given appropriate credit. Contents may not be used for commercial purposes, or adapted, remixed, transformed or built upon. (https://creativecommons.org/licenses/by-nc-nd/4.0/).

Georg Thieme Verlag KG
Stuttgart · New York