Diagnosis of liver cancer by FTIR spectra of serum

Xien Yang; Quanhong Ou; Weiye Yang; Youming Shi; Gang Liu

doi:10.1016/j.saa.2021.120181

Diagnosis of liver cancer by FTIR spectra of serum

Spectrochim Acta A Mol Biomol Spectrosc. 2021 Dec 15:263:120181. doi: 10.1016/j.saa.2021.120181. Epub 2021 Jul 17.

Authors

Xien Yang¹, Quanhong Ou¹, Weiye Yang², Youming Shi³, Gang Liu⁴

Affiliations

¹ School of Physics and Electronic Information, Yunnan Normal University, Kunming 650500, China.
² School of Preclinical Medicine, Zunyi Medical University, Zunyi 563003, China.
³ School of Physics and Electronic Engineering, Qujing Normal University, Qujing 655011, China.
⁴ School of Physics and Electronic Information, Yunnan Normal University, Kunming 650500, China. Electronic address: gliu66@163.com.

PMID: 34311164
DOI: 10.1016/j.saa.2021.120181

Abstract

Liver cancer is the most common fatal malignant tumor in the world. Early diagnosis of liver cancer can improve the survival rate of the patients with liver disease. In this paper, Fourier transform infrared (FTIR) spectroscopy combined with curve fitting and chemometrics was used to distinguish the serum from patients from that of healthy people. The curve fitting results in protein range of 1700-1600 cm^-1 showed that there were differences in the secondary structure of protein in serum between the patients with liver cancer and healthy people. Principal component analysis (PCA) in lipid range of 2900-2800 cm^-1 could distinguish the serum of patients with liver cancer from that of healthy people. The first two principal components PC1 and PC2 explained 95% of the total data variance. The sensitivity and specificity of partial least squares discriminant analysis (PLS-DA) in lipid range of 2900-2800 cm^-1 reached 92.85% and 95.23% respectively. It is shown that FTIR spectroscopy might be developed as an effective method for the diagnosis of liver cancer.

Keywords: Curve fitting; FTIR spectroscopy; Liver cancer; Partial least squares discriminant analysis; Principal component analysis; Serum.

MeSH terms

Discriminant Analysis
Fourier Analysis
Humans
Least-Squares Analysis
Liver Neoplasms* / diagnosis
Principal Component Analysis
Spectroscopy, Fourier Transform Infrared