Optical biopsy of skin cancer based on Raman and fluorescence spectroscopy

This paper presents the application of the optical methods of Raman spectroscopy and autofluorescence analysis for studying of the skin diagnostic features. Benign and malignant skin tumors from 89 patients were measured using portable spectroscopic system with near-infrared 785 nm laser excitation. Classifications of the different skin tumors were made using partial least squares methods with discriminant analysis in two tasks: melanoma from other malignant skin tumors (basal cell carcinoma and squamous cell carcinoma) with 0.95 sensitivity and 0.86 specificity and melanoma from pigmented nevus with 0.68 sensitivity and 0.78 specificity. The main findings establish that combined analysis of the autofluorescence and Raman spectral signatures of the different skin tumors are more effective in comparison with using of the only Raman spectroscopy. Also, results of classifications show the tumor spectra contain most useful information rather than tumor spectra normalized to the corresponding normal skin spectra.