Spectral Features Selection and Classification for Bimodal Optical Spectroscopy Applied to Bladder Cancer In Vivo Diagnosis

Author

Pery, Emilie ; Blondel, Walter C. P. M. ; Tindel, Samy ; Ghribi, Maha ; Leroux, Agnes ; Guillemin, Francois

Author_Institution

Image Sci. for Interventional Tech. Lab. (ISIT), Auvergne Univ., Clermont-Ferrand, France

Volume

61

Issue

1

fYear

2014

fDate

Jan. 2014

Firstpage

207

Lastpage

216

Abstract

This paper describes an experimental study combining spatially resolved autofluorescence (AF) and diffuse reflectance (DR) fibred spectroscopies to discriminate in vivo between healthy and pathological tissues in a preclinical model of bladder cancer. Then, a detailed step-by-step analysis scheme is presented for the extraction and the selection of discriminative spectral features (correlation, linear discriminant, and logistic regression analysis), and for the spectroscopic data final classification algorithms (regularized discriminant analysis and support vector machines). Significant differences between healthy, inflammatory, and tumoral tissues were obtained by selecting a reasonable number of discriminant spectral features from AF, DR, and intrinsic fluorescence spectra, leading to improved sensitivity (87%) and specificity (77%) compared to monomodality (AF or DR alone).

Keywords

cancer; correlation methods; feature extraction; feature selection; fluorescence; medical signal processing; patient diagnosis; physiological models; regression analysis; signal classification; support vector machines; tumours; AF; DR; autofluorescence; bimodal optical spectroscopy; bladder cancer in vivo diagnosis; correlation analysis; diffuse reflectance fibred spectroscopies; discriminative spectral feature extraction; discriminative spectral feature selection; healthy tissue; inflammatory tissue; intrinsic fluorescence spectra; linear discriminant analysis; logistic regression analysis; monomodality; pathological tissues; preclinical model; regularized discriminant analysis; spectroscopic data final classification algorithm; step-by-step analysis scheme; support vector machines; tumoral tissue; Absorption; Bladder; Cancer; Feature extraction; In vivo; Spectroscopy; Tumors; Autofluorescence (AF); bladder cancer; diffuse reflectance (DR); feature selection; supervised classification;

fLanguage

English

Journal_Title

Biomedical Engineering, IEEE Transactions on

Publisher

ieee

ISSN

0018-9294

Type

jour

DOI

10.1109/TBME.2010.2103559

Filename

5680600