Spectra Measurement And Analysis Of Endogenous Biomarks For Colonic Adenocarcinoma Ex Vivo

Autofluorescence spectroscopy is a promising tool for the diagnosis of early colorectal cancer because of its non-invasive and real-time visualization with high sensitivity. However, the existing autofluorescence imaging systems for early diagnosis of colorectal cancer are still in the preliminary clinical application due to the low specificity. In the theses, the recent advances in autofluorescence spectroscopy and imaging techniques for early diagnosis of colorectal cancer are reviewed. Secondly, under four different optimal excitation wavelengths (337,375,405and460nm), AF spectra of freshly excised normal and adenocarcinoma colon tissues were measured. Pattern recognition method including features extraction, data reduction using principle component analysis (PCA) and Fisher’s discriminant analysis (FDA) were performed for spectral data classification. The results suggest that the pattern recognition of the multiple excitation AF spectra is an effective algorithm for improving the diagnostic accuracy of adenocarcinoma. Thirdly, autofluorescence spectra of colonic normal and adenocarcinoma tissues under337nm excitation were analyzed using multivariate curve resolution alternating least squares (MCR-ALS) with non-negativity constraint. The results imply that MCR-ALS is a powerful tool for characterizing the spectra profiles of the main endogenous biomarks in neoplasm transformation. Collagen, NAD(P)H and elastin were identified as the main contributing endogenous biomarks. Fourthly, ex vivo colorectal adenocarcinoma samples which exceptionally exhibited635nm fluorescence emission under405nm excitation were further studied to explore the correlation between porphyrins and colonic adenocarcinoma. The spectral profile and lifetime of the red fluorescence are similar to that of porphyrins, which indicates that635nm fluorescence arise from porphyrins fluorescence. Finally, Monte Carlo simulation was applied to analyze the light distribution of the converging light beam in the homogenous colorectal tissue model. The optimal choice of the depth of focus, detection area and source-detector distance for the diffuse reflectance spectroscopy measurement were also analyzed. The results show that the optimal focus depth of the non-contact spectrum system for diagnosis of early colorectal cancer is1mm. For the diffuse reflectance spectroscopy measurement, the source-detector distance can be determined by the intensity of emission light and the response sensitivity of detector. Compared with0.1mm,0.6mm is more appropriate detection radius for the diffuse reflectance spectroscopy measurement. In order to reduce the influence on the spectrum caused by the change of detection depth, the apex angle of detection light cone should be constant.In conclusion, in order to improve the diagnosis accuracy of early colorectal cancer using autofluorescence technique, the difference between the autofluorescence spectra of human colorectal normal and adenocarcinoma tissue and its originations were investigated. The diagnosis algorithm with higher accuracy was established to classify different types of colonic tissue from the measured autofluorescence spectra. Moreover, the light distribution of the converging light beam in the colonic tissue and the optimal choice of the depth of focus, detection area and source-detector distance for the diffuse reflectance spectroscopy measurement were also analyzed for establishing non-contact spectroscopy detection system.