Bimodal spectroscopic evaluation of ultra violet-irradiated mouse skin inflammatory and precancerous stages: instrumentation, spectral feature extraction/selection and classification (k-NN, LDA and SVM)

Abstract

et al.. Bimodal spectroscopic evaluation of ultra violet-irradiated mouse skin inflammatory and precancerous stages: instrumentation, spectral feature extraction/selection and classification (k-NN, LDA and SVM).Abstract. 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 pre-clinical 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 Sensibility (Se) and Specificity (Sp) as a function of the selected features, of the combinations of 3 different inter-fibres 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.

Intradermal tacrolimus prevent scar hypertrophy in a rabbit ear model: a clinical, histological and spectroscopical analysis

“…Our spatially resolved bimodal spectroscopy system has already been described in detail elsewhere (20)(21)(22). In the present study, we used seven AF excitation peaks ranging from 360 to 440 nm and a wide wavelength band (330-720 nm) for DR.…”

Hybrid feature selection and SVM-based classification for mouse skin precancerous stages diagnosis from bimodal spectroscopy

“…Therefore, fusing the two types of optical information gets a wider insight into the skin modi-fications. In a second published study [20], our team showed how spatial resolution (the use of several distances between excitation and collection optical fibers) increases diagnosis accuracy as well as the fact that SVM is the most appropriate classification algorithm to our problem at hand (still in a pairwise approach with a priori knowledge on the data) acquired through a bimodal approach. In order to get closer to medical interest, we propose in the current study to process data obtained through a spatially-resolved bimodal approach first, with no a priori knowledge of the data set (no extraction of spectral features based on spectra visualization) and second, to base our SVM-classification on a multi-class approach (instead of a decomposition scheme).…”

Multidimensional spectroscopic data fusion improves precancerous tissue discrimination based on spatially resolved autofluorescence and diffuse reflectance spectroscopy

“…The validation of our approach is based on a preprocessed spectroscopic data set acquired on mice skin healthy and hyperplastic and dysplastic tissues in the framework of a preclinical UV-induced carcinogenesis study already presented in details together with the experimental set-up in Amouroux et al 1 and Diaz et al 16 . This latter data set corresponds to a multidimensional matrix containing a total of 252 tissue sites, each site being characterized by (i) seven AF intensity spectra (corresponding to seven narrow-band wavelength excitations) and shaped into Emission-Excitation Matrices (EEM), (ii) one DR spectrum (obtained for a large-band wavelength illumination) and (iii) the histological class determined by the histopathological analysis (considered as gold standard classification).…”

“…Our works are focused on the combination of steady-state and spatially resolved Diffuse Reflectance (DR) spectroscopy and multiple excitation AutoFluorescence (AF) spectroscopy to improve diagnosis efficiency of skin precancerous tissues 1,10,11 . For simplicity and cost effective instrumentation considerations, steadystate excitation and spatial resolution are preferred 16 . A multiple optical fiber-probe is then used in which several collecting fibers at different distances from one excitation fiber (CEFS, Collection to Excitation Fiber Separation) are selected.…”

“…precancerous) modifications arise in the epidermis 1 . From the multidimensional spectroscopic measures acquired with these systems (wavelength resolved spectra, at several CEFS, AF excitations and modalities), discriminant features need to be extracted in order to best classify different precancerous states of skin tissues 1,16 . One of the difficulty is for discriminating between states showing high "confusion" between them like for instance atypical lesions 2 (that have a potential risk of progressing towards invasive lesions) and dysplasia (that is close to in situ cancer, most at risk at evolving towards invasive cancer).…”