Automatic and objective oral cancer diagnosis by Raman spectroscopic detection of keratin with multivariate curve resolution analysis

Po Hsiung Chen, Rintaro Shimada, Sohshi Yabumoto, Hajime Okajima, Masahiro Ando, Chiou Tzu Chang, Li Tzu Lee, Yong Kie Wong, Arthur Chiou, Hiro O. Hamaguchi

Research output: Contribution to journalArticle

20 Citations (Scopus)

Abstract

We have developed an automatic and objective method for detecting human oral squamous cell carcinoma (OSCC) tissues with Raman microspectroscopy. We measure 196 independent Raman spectra from 196 different points of one oral tissue sample and globally analyze these spectra using a Multivariate Curve Resolution (MCR) analysis. Discrimination of OSCC tissues is automatically and objectively made by spectral matching comparison of the MCR decomposed Raman spectra and the standard Raman spectrum of keratin, a well-established molecular marker of OSCC. We use a total of 24 tissue samples, 10 OSCC and 10 normal tissues from the same 10 patients, 3 OSCC and 1 normal tissues from different patients. Following the newly developed protocol presented here, we have been able to detect OSCC tissues with 77 to 92% sensitivity (depending on how to define positivity) and 100% specificity. The present approach lends itself to a reliable clinical diagnosis of OSCC substantiated by the 'molecular fingerprint' of keratin.

Original languageEnglish
Article number20097
JournalScientific Reports
Volume6
DOIs
Publication statusPublished - 2016 Jan 25

    Fingerprint

ASJC Scopus subject areas

  • General

Cite this

Chen, P. H., Shimada, R., Yabumoto, S., Okajima, H., Ando, M., Chang, C. T., Lee, L. T., Wong, Y. K., Chiou, A., & Hamaguchi, H. O. (2016). Automatic and objective oral cancer diagnosis by Raman spectroscopic detection of keratin with multivariate curve resolution analysis. Scientific Reports, 6, [20097]. https://doi.org/10.1038/srep20097