Hyperspectral microscopy and cluster analysis for oral cancer diagnosis

Research output: Chapter in Book/Report/Conference proceedingPoster abstractpeer-review

3 Citations (Scopus)

Abstract

Oral cancer incidences have been increasing in recent years and late detection often leads to poor prognosis. Raman spectroscopy has been identified has a valuable diagnostic tool for cancer but its time consuming nature has prevented its clinical use. For Raman to become a realistic aid to histopathology, a rapid pre-screening technique is required to find small regions of interest on tissue sections [1]. The aim of this work is to investigate the feasibility of hyperspectral imaging in the visible spectral range as a fast imaging technique before Raman is performed. We have built a hyperspectral microscope which captures 300 focused and intensity corrected images with wavelength ranging from 450-750 nm in around 30 minutes with sub-micron spatial resolution and around 10 nm spectral resolution. Hyperstacks of known absorbing samples, including fluorescent dyes and dried blood droplets, show excellent results with spectrally accurate transmission spectra and concentration-dependent intensity variations. We successfully showed the presence of different components from a non-absorbent saliva droplet sample. Data analysis is the greatest hurdle to the interpretation of more complex data such as unstained tissue sections.

Original languageEnglish
Title of host publicationHigh-Speed Biomedical Imaging and Spectroscopy: Toward Big Data Instrumentation and Management II
PublisherSPIE
Volume10076
ISBN (Electronic)9781510605930
DOIs
Publication statusPublished - 2017
EventHigh-Speed Biomedical Imaging and Spectroscopy: Toward Big Data Instrumentation and Management II 2017 - San Francisco, United States
Duration: 30 Jan 20171 Feb 2017

Conference

ConferenceHigh-Speed Biomedical Imaging and Spectroscopy: Toward Big Data Instrumentation and Management II 2017
Country/TerritoryUnited States
CitySan Francisco
Period30/01/20171/02/2017

Keywords

  • classification
  • cluster analysis
  • diagnosis
  • hyperspectral
  • microscopy
  • oral cancer
  • Raman spectroscopy
  • transmission

Fingerprint

Dive into the research topics of 'Hyperspectral microscopy and cluster analysis for oral cancer diagnosis'. Together they form a unique fingerprint.

Cite this