Improving colorectal cancer detection by extending the near-infrared wavelength range and tissue probed depth of diffuse reflectance spectroscopy: A support vector machine approach

Research output: Chapter in Book/Report/Conference proceedingsChapterpeer-review

Abstract

Colorectal cancer (CRC) is the second most deadly and third most common type of cancer worldwide. In this study, we assessed the improvement of the diagnostic potential of diffuse reflectance spectroscopy (DRS) for CRC detection upon extending the tissue probed depth (up to 2mm) and wavelength ranges (350-1919 nm) investigated in previous studies. We analyzed almost 3000 DR spectra (7.5 times more than previous studies) collected with 630-?m and 2500-?m source-detector distance probes by using support vector machines with potential to automate tissue classification. We achieved 96.1% sensitivity and 95.7% specificity and 0.987±0.005 AUC on tissue classification.

Original languageEnglish
Title of host publicationOptical Biopsy XX
Subtitle of host publicationToward Real-Time Spectroscopic Imaging and Diagnosis
EditorsRobert R. Alfano, Stavros G. Demos, Angela B. Seddon
PublisherSPIE
ISBN (Electronic)9781510647794
DOIs
Publication statusPublished - 2022
EventOptical Biopsy XX: Toward Real-Time Spectroscopic Imaging and Diagnosis 2022 - Virtual, Online
Duration: 20 Feb 202224 Feb 2022

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume11954
ISSN (Print)1605-7422

Conference

ConferenceOptical Biopsy XX: Toward Real-Time Spectroscopic Imaging and Diagnosis 2022
CityVirtual, Online
Period20/02/2224/02/22

Keywords

  • Artificial intelligence
  • Biomedical optics
  • Biophotonics
  • Colon cancer
  • Colonoscopy
  • Colorectal cancer
  • Diffuse reflectance spectroscopy
  • Elastic scattering spectroscopy
  • Hyperspectral imaging
  • Machine learning
  • Near-infrared spectroscopy
  • Optical diagnostics
  • Optical spectroscopy
  • Surgical guidance

Fingerprint

Dive into the research topics of 'Improving colorectal cancer detection by extending the near-infrared wavelength range and tissue probed depth of diffuse reflectance spectroscopy: A support vector machine approach'. Together they form a unique fingerprint.

Cite this