Vibrational Spectroscopy and Cancer - Vibrational Spectroscopy for Tissue Analysis - page 90

a band at 991 cm −1 , which is weak in colon tissues; and (b) the packing and
degree of disorder of membrane lipids were close to those observed in nor-
mal colonic tissues. They concluded three important aspects from their
study: (i) IR spectroscopy can be used in combination with pressure tuning
as a useful method to address problems of tumour biology in cell culture
systems, (ii) cell lines offer a useful experimental model to explore the origin
of the spectroscopic changes that were observed in colon cancer tissues, and
(iii) the malignant colonocyte is the likely source of all or most spectroscopic
abnormalities of human colon cancer [155].
Pinzaru et al. reported for the first time on the application of surface-
enhanced Raman scattering (SERS) on normal and altered epithelial layer
in human colon carcinoma tissues [156]. It was demonstrated that different
tissue structures of tumour and normal colon have characteristic features in
SERS spectra.
Widjaja et al. combined NIR Raman spectroscopy (785-nm laser excitation)
with support vector machines (SVMs) for improving multiclass classifica-
tion among different histopathological groups in tissues [157]. A total of 105
colonic tissue specimens from 59 patients including 41 normal, 18 hyperplas-
tic polyps, and 46 adenocarcinomas were used for this purpose. A total of
817 tissue Raman spectra were acquired and subjected to PCA, in which 324
Raman spectra were from normal, 184 from polyps, and 300 from adenocar-
cinomatous colonic tissue. Two types of SVMs (i.e., C-SVM and v-SVM) with
three different kernel functions (linear, polynomial, and Gaussian radial
basis function (RBF) in combination with PCA was used to develop effective
diagnostic algorithms for classification of Raman spectra of different colonic
tissues. The results showed that using different methods, a diagnostic accu-
racy of 98.5-99.9% can be achieved. It was also shown that all the polyps can
be identified as normal and adenocarcinomatous.
Kondepati et al. used FTIR spectroscopy for the detection of structural dis-
orders in colorectal cancer DNA [158]. The DNA was isolated from cancer
and its adjacent histological normal tissue of 43 colorectal cancer patients.
Spectral differences among grade 1, 2, and 3 cancers were observed. Minor
structural disorders in cancer DNA were also detected by differences in
the spectral regions assigned to nucleotide bases and the phosphodiester-
deoxyribose backbone. Using linear discriminant analysis, diagnostic
accuracies between 70% and 86% were achieved using different validation
strategies, the lowest achieved by using a variable selection scheme indepen-
dent from the validation test data.
Li et al. reported on application of ATR-FTIR spectroscopy and fibre-
optic technology for detecting in vivo and in situ colorectal cancer [159].
In this study, a total of five patients with large intestine cancer were detected
in vivo and in situ. Of them, three cases of colon cancer and one case of
cecum cancer, were detected intra-operatively and in vivo by using an FTIR
spectrometer during surgical operation, and one case of rectum cancer was
explored noninvasively and in vivo before the surgical operation. It was