Data mining methods of lung cancer diagnosis by saliva tests using surface enhanced Raman spectroscopy

Surface Enhanced Raman Spectroscopy (SERS) is a trace amount substance detecting technique developing quickly in recent years. In this paper, the saliva SERS spectrum of 59 lung cancer patients and 18 normal people were measured, and analyzed with data mining technology and the traditional statistical classification methods. The data were established by the Support Vector Machine (SVM), Random Forests algorithm (RF) and Fisher discriminant model, and discussed the auxiliary diagnosis efficiency for lung cancer with the models. The diagnosis indexes of the SVM and RF algorithm are higher than Fisher discriminant analysis, and it can be thought that they are judging the optimal classification model of lung cancer. Compared with the healthy people, the results show that the study on diagnosis of the lung cancer by SERS on data mining can be a new type of the lung cancer diagnosis tool.