This paper deals with the development and application of in vivo spatially-resolved bimodal spectroscopy
(AutoFluorescence AF and Diffuse Reflectance DR), to discriminate various stages of skin precancer in a preclinical
model (UV-irradiated mouse): Compensatory Hyperplasia CH, Atypical Hyperplasia AH and Dysplasia D. A programmable
instrumentation was developed for acquiring AF emission spectra using 7 excitation wavelengths:
360, 368, 390, 400, 410, 420 and 430 nm, and DR spectra in the 390–720 nm wavelength range. After various steps of intensity
spectra preprocessing (filtering, spectral correction and intensity normalization), several sets of spectral
characteristics were extracted and selected based on their discrimination power statistically tested for
every pair-wise comparison of histological classes. Data reduction with Principal Components Analysis (PCA)
was performed and 3 classification methods were implemented (k-NN, LDA and SVM), in order to compare diagnostic
performance of each method. Diagnostic performance was studied and assessed in terms of sensitivity (Se) and
specificity (Sp) as a function of the selected features, of the combinations of 3 different inter-fibers distances and of the
numbers of principal components, such that: Se and Sp≈ 100% when discriminating CH vs. others; Sp≈ 100% and Se> 95% when discriminating Healthy vs. AH or D; Sp≈ 74% and Se≈ 63% for AH vs. D.