TITLE:
Raman spectroscopy for human cancer tissue diagnosis: A pattern recognition approach
AUTHORS:
Maher Rizkalla, Parvin Ghane, Mangilal Agarwal, Sudhir Shrestha, Kody Varahramyan
KEYWORDS:
Cancer; Tissues; Raman Spectroscopy; Pattern Recognition
JOURNAL NAME:
Journal of Biomedical Science and Engineering,
Vol.5 No.12A,
December
31,
2012
ABSTRACT: In
this work, optical scattering using Raman spectroscopy has been analyzed for
various cancer tissues. The Raman shifts obtained at the Indiana University
Bloomington (IUB) and Indiana University-Purdue University Indianapolis (IUPUI)
laboratories have been processed for diagnosing various types of cancer
tissues. The objective of this research is to distinguish between cancerous and
non-cancerous tissues. Small size tissue samples have been processed, seeking
the minimum size tissue that can be diagnosed via Raman spectroscopy. The tests have been conducted on nearly 20 human
tissues. A Matlab program has been written following Parzen-Window classifier
to recognize the Raman shift pattern for various types of cancer tissues,
including breast cancer, kidney, and Gyn-Uterus. A software visual model has
been used for data processing. Unique signals for breast and kidney tumors
have been obtained. The approach followed in this paper shows promise for early
cancer detection in humans.