Near infrared hyperspectral imaging to evaluate tongue tumor resection margins intraoperatively (Conference Presentation)

Susan G. Brouwer de Koning; M.B. Karakullukcu; L. Smit; E.J.M. Baltussen; H.J.C.M. Sterenborg; T.J.M. Ruers

doi:10.1117/12.2289166

14 March 2018 Near infrared hyperspectral imaging to evaluate tongue tumor resection margins intraoperatively (Conference Presentation)

Susan G. Brouwer de Koning, M.B. Karakullukcu, L. Smit, E.J.M. Baltussen, H.J.C.M. Sterenborg, T.J.M. Ruers

Author Affiliations +

Proceedings Volume 10469, Optical Imaging, Therapeutics, and Advanced Technology in Head and Neck Surgery and Otolaryngology 2018; 104690G (2018) https://doi.org/10.1117/12.2289166
Event: SPIE BiOS, 2018, San Francisco, California, United States

Abstract

Positive tumor resection margins are reported in up to 45% of the patients undergoing surgery for tongue cancer. With the aim to develop a technique that can assess tumor resection margins intraoperatively, we conducted an ex vivo study to evaluate the feasibility of near infrared hyperspectral imaging for distinguishing tumor from healthy tongue tissue. Fresh surgical specimens of squamous cell carcinoma of the tongue were scanned with a pushbroom camera. The acquired spectral hypercubes contain a measure of the diffuse light reflectance (wavelength range of 900-1700 nm) for each pixel of the hyperspectral image. Spectral bands were selected from the spectrum and used to classify spectra of tumor and healthy tissue. In this, a linear classifier was trained on 80% of the data and its performance in predicting the tissue type of the residual 20% of the data was measured. This was repeated five times and mean accuracy, sensitivity and specificity were used as output for this study. A total of 463 spectra were obtained from tongue tumor tissue and 421 spectra from healthy tongue tissue. The spectral bands between 1060-1130 nm and 1150-1190 nm were used in the classification analysis. Mean accuracy, sensitivity and specificity were 89%±13, 94%±11 and 87%±21, respectively. Near infrared hyperspectral imaging can discriminate tongue tumor tissue from healthy tongue tissue in an ex vivo setting by using specific bands of the reflectance spectrum. Further analyses will be done to assess whether using the whole spectrum can improve the classification results.

Conference Presentation

Citation Download Citation

Susan G. Brouwer de Koning, M.B. Karakullukcu, L. Smit, E.J.M. Baltussen, H.J.C.M. Sterenborg, and T.J.M. Ruers "Near infrared hyperspectral imaging to evaluate tongue tumor resection margins intraoperatively (Conference Presentation)", Proc. SPIE 10469, Optical Imaging, Therapeutics, and Advanced Technology in Head and Neck Surgery and Otolaryngology 2018, 104690G (14 March 2018); https://doi.org/10.1117/12.2289166

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available