Raman Spectroscopy: A Novel Technology for Gastric Cancer Diagnosis

Abstract

Gastric cancer is usually diagnosed at late stage and has a high mortality rate, whereas early detection of gastric cancer could bring a better prognosis. Conventional gastric cancer diagnostic methods suffer from long diagnostic times, severe trauma, and a high rate of misdiagnosis and rely heavily on doctors’ subjective experience. Raman spectroscopy is a label-free molecular vibrational spectroscopy technique that identifies the molecular fingerprint of various samples based on the inelastic scattering of monochromatic light. Because of its advantages of non-destructive, rapid, and accurate detection, Raman spectroscopy has been widely studied for benign and malignant tumor differentiation, tumor subtype classification, and section pathology diagnosis. This paper reviews the applications of Raman spectroscopy for the in vivo and in vitro diagnosis of gastric cancer, methodology related to the spectroscopy data analysis, and presents the limitations of the technique.

Raman spectroscopy: A prospective intraoperative visualization technique for gliomas

“…Since the composition of nucleic acids, proteins and lipids in tumor cells are dramatically different from normal brain cells, while Raman spectroscopy is able to identify the chemical composition of a sample by measuring the Raman shift caused by the difference between the frequency of the scattered light and incident light in the Raman effect ( 33 ), as Ji reported that glioma and normal brain tissues have significantly different spectral peaks at 1080 cm-1 (nucleic acid), 2845 cm-1 (lipid) and 2930 cm-1 (protein) positions ( 34 ). Instead, the differences just depend on their chemical group, independent of the excitation wavelength, while the intensity of Raman peaks is related to the excitation light wavelength, power and concentration of the measured substance ( 35 ), so the Raman spectroscopy have the possibility to evaluate tumors and normal tissues according to the differences ( Figure 1 ) ( 12 ). Currently, the major diagnostic methods for glioma in clinical practice include imaging and pathology, and imaging primarily covers CT, MRI and PET.…”

Building an ensemble learning model for gastric cancer cell line classification via rapid raman spectroscopy

“…Relevant studies have shown that analysing the fingerprint region alone is superior to analysing the HW region alone, but analysing the fingerprint and HW regions together is optimal [17] , [18] . In addition, in previous Raman spectroscopy studies, spectral baselines were removed in the spectral preprocessing stage using methods such as adaptive iteratively reweighted penalized least squares (airPLS) algorithm [19] , [20] , [21] . However, Tomasz et al speculated in the study that there is a correlation between the Raman baseline and the fluorescence background of the sample [22] .…”

“…Due to the small characteristic differences among Raman spectra, computer methods such as machine learning and deep learning are often used to differentiate the spectra when analysing them. Algorithms such as the principal component analysis-linear discriminant analysis (PCA-LDA), K-nearest neighbour (KNN), partial least squares-discriminant analysis (PLS-DA), support vector machine (SVM), artificial neural network (ANN), and convolutional neural network (CNN) algorithms are widely used [20] , [26] , [27] . PCA can perform linear dimensionality reduction by orthogonal transformation [28] and reduce the difficulty of data analysis.…”

Identification of Uric Acid Crystals Accumulation in Human and Animal Tissues Using Combined Morphological and Raman Spectroscopy Analysis

“…Raman spectroscopy is a nondestructive and effective tool that has been applied successfully to various fields such as: medicine [ 19 , 20 , 21 , 22 , 23 ], dental research [ 24 , 25 , 26 ], biology [ 27 ], food safety [ 28 ], chemistry [ 29 ], pharmacy [ 30 ], etc. Material identification and its structural characterization are complex, since Raman scattering acts at various levels, from surface to the depth of the samples.…”