Translator Disclaimer
22 February 2017 Hyperspectral microscopy and cluster analysis for oral cancer diagnosis
Author Affiliations +
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.
© (2017) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Anneliese Jarman, Arunthathi Manickavasagam, Neveen Hosny, and Frederic Festy "Hyperspectral microscopy and cluster analysis for oral cancer diagnosis", Proc. SPIE 10076, High-Speed Biomedical Imaging and Spectroscopy: Toward Big Data Instrumentation and Management II, 100761I (22 February 2017);

Back to Top