The Study On Diagnosis Of Gastric Cancer With Raman Spectroscopy

Spectroscopy analysis technique is efficient, rapid, accurate, low-cost. It have a profound affection on the clinical medical detection. According to laser-induced Raman spectra can reflect the molecular structure information, and the changes of the microenvironment. The purpose of this paper is to research a auxiliary method of early diagnosis of gastric cancer by Raman spectrum technology when the serum is used as sample.In this paper, the Raman and fluorescence spectra of twenty normal cases, thirty gastric cancer patients and twenty atrophic gastritis patients were excited by Ar-ion laser (514.5nm and 488.0nm), and collected in the range of 500nm-620nm and 520nm-640nm. The principal component analysis (PCA) and discriminant analysis of the Raman spectra was used for analysis and data processing creatively.From the statistic results of Raman spectra and its second derivative spectra we could obtain several conclusions. The serum spectum is different among normal, atrophic gastritis and gastric cancer. The position of the negative peaks of second derivative spectra are almost the same, and the normal cases'derivative spectra were the most steep, the gastric cancer's were the smoothest of all. We introduce a parameterΔλas the red shift of fluorescence peak, a notable differerce is thatΔλis often bigger than 13nm for the atrophic gastritis and gastric cancer less than 13nm for normal cases. Then we introduce a parameterμas the ratio of Raman relative intensity. We found that parameterμof the atrophic gastritis is bigger than 1.0, but less than 1.0 for the gastric cancer. We considered the parameterμas a criterion and got an accuracy of 81.8%, a sensitivity of 80%, a specificity of 82.9% for distinguish of atrophic gastritis and gastric cancer. Then we analyzed the PCA of fifty gastric cancer's and atrophic gastritis'serum samples by SPSS software and got score scatterplots of Raman spectroscopy. Using the scatterplots we were able to distinguish samples easily and got an accuracy of 92%, a sensitivity of 90%, a specificity of 95%. And this result the same as discriminant analysis in theory. This analysis method has a higher rate of accurate.It may be useful in cancer diagnosis, treatment and early prevention.We analyzed the mechanism and obtained that the differences of Raman spectroscopy mainly affected by (3-Carotene; Flavins, hematoporphyrin derivatives, zinc porphyrin and the change of serum tissue microenvironment may cause the red shift.