Study On Smears Of Nasopharyngeal Carcinoma Tissue Raman Spectroscopy

Nasopharyngeal carcinoma (NPC) is one of malignant tumors in the head and neck that threats human life and health. In this study, in order to overcome the shortcoming of destructive detection and bleeding during NPC biopsy, we explore a new method for discriminating nasopharyngeal carcinoma from non-cancerous smears based on Raman spectroscopy and SERS technique. Multivariate statistical analysis were used to develop diagnostic algorithms for differentiation of NPC from non-cancerous smears. In addition, we discuss Raman spectral differences of two methods between nasopahryngela smear and tissue in this preliminary study.1.Using nasopharyngeal smear to study 39 NPC and non-cancerous samples based on Raman spectroscopy technique. The Principal components analysis (PCA) and linear discriminant analysis (LDA) algorithm achieved a diagnostic sensitivity of 82.1% and specificity of 82.9% for differentiating NPC from nasopharyngeal non-cancerous smears. On the other hand, we explore average Raman spectrum and pathological images about different types of nasopharyngeal smears and analyze the Raman spectral shape, the special peak position, the intensity of Raman spectroscopy, and the images of cell morphology.2.As the SERS-active nanostructures, Ag nanoparticles were directly dropped on NPC and non-cancerous smears to enhance the Raman scattering signal. The PCA-LDA algorithm provides the sensitivity of 79.4% and specificity of 81.8% for differentiation between cancerous and non-cancerous nasopharyngeal smears. The results show that the method has the ability for discriminating cancerous form non-cancerous nasopharyngeal smears by SERS technique.3.The methods of nasopharyngeal smear and tissue are used to discuss the average Raman spectra in the fingerprint (400-1800cm-1) and high wavenumber (2800-3100cm-1) regions. The preliminary results show that the nasopharyngeal smear method has a improvement in the non-destructive treatment than nasopharyngeal tissue. Furthermore, the multivariate statistical method is used for discriminating nasopharyngeal non-cancerous from NPC tissue, and generates the sensitivities of 95.4% and specificities of 91.9%. This study demonstrates that the high wavenumber Raman spectroscopy has the potential for diagnosis and detection in early nasopharyngeal carcinoma.