Multivariate discriminating algorithm for analyzing laser-induced fluorescence spectra of human gastrointestinal cancer

The purpose of this study has been to evaluate the laser- induced fluorescence characters of normal and malignant stomach tissue in Vitro and in Vivo. The stepwise multivariate discrimination analysis was used to make a multivariate statistical algorithm for analyzing the diagnostic parameters of human stomach tissues fluorescence spectrum. The resulting spectra could be differentiating histologically stomach abnormal tissue from normal tissue with a sensitivity and specificity value of 95% and 97%. The diagnosis results were in excellent agreement with histopathological results.