Discrimination of lung cancer by serum using fluorescence and principal component analysis

Su Zhang, Xiaozhou Li, Tianyue

Abstract

The technology of laser-induced auto-fluorescence spectroscopy was used on serum for the diagnosis of lung cancer. Serum from 30 lung cancer patients and 20 healthy people were collected and measured. The results have shown that there was significant difference in the fluorescence spectroscopy between those from lung cancer patients and the controls. Then, we use principal component analysis and discriminant analysis to analyze spectra, and got an accuracy of 88% in distinguishing lung cancer patients and healthy people. Our experiment revealed that fluorescence of serum can be an indicator for the diagnosis of lung cancer.