Different classification algorithms and serum surface enhanced Raman spectroscopy for noninvasive discrimination of gastric diseases

• DOI: 10.1002/jrs.4924
• Publication Date: 2016-04-10T18:12:24Z
• Authors: Xiaozhou Li; Tianyue Yang; Siqi Li; Deli Wang; Youtao Song; Kedong Yu
• Keywords: Surface-Enhanced Raman Spectroscopy

In this study, surface enhanced Raman spectroscopy (SERS) was used to investigate the spectral characteristics of blood serum for purpose diagnosing stomach diseases. SERS data collected from patients with atrophic gastritis, both pre‐operation and post‐operation gastric cancer, healthy individuals. Visual differences in spectra were observed between four groups which indicate corresponding biomolecule concentration changes blood. To further diagnostic ability human serum, analyzed three chemometric processes. These methods extracted features classified data. Principal component analysis (PCA) first performed reduce dimensionality original Then, classification support vector machine (SVM), linear discriminant (LDA) regression tree (CART) evaluation ability. Accuracies 96.5%, 88.8% 87.1% obtained PCA‐SVM, PCA‐LDA PCA‐CART, respectively. Copyright © 2016 John Wiley & Sons, Ltd.