Raman spectral post-processing for oral tissue discrimination - a step for an automatized diagnostic system

Author(s)

Carvalho, Luis Felipe C. S. ^[1] ; Nogueira, Marcelo Saito ^[2] ; Neto, Lazaro P. M. ^[1] ; Bhattacharjee, Tanmoy T. ^[1] ; Martin, Airton A. ^{[3, 4]}

Affiliation

^[1] Univap, Inst Pesquisa & Desenvolvimento, Lab Espectroscopia Vibrac Biomed, Ave Shishima Hifumi 2911, BR-12244000 Sao Jose Dos Campos, SP - Brazil

^[2] Univ Sao Paulo, Sao Carlos Inst Phys, Opt Grp, Biophoton Div, Ave Trabolhador Sao Carlense 400, BR-13566590 Sao Carlos, SP - Brazil

^[3] Univ Brasil UnBr, Biomed Vibrat Spect Grp, Biomed Engn Innovat Ctr, Rua Carolina Fonseca 235, BR-08230030 Sao Paulo, SP - Brazil

^[4] Univ Fed Piaui UFPI, Dept Fis CCN Bairro Ininga, Campus Ministro Petronio Portella, BR-64049550 Teresina, PI - Brazil

• Document type: Journal article
• Source: BIOMEDICAL OPTICS EXPRESS; v. 8, n. 11, p. 5218-5227, NOV 1 2017.
• Web of Science Citations: 10

Abstract

Most oral injuries are diagnosed by histopathological analysis of a biopsy, which is an invasive procedure and does not give immediate results. On the other hand, Raman spectroscopy is a real time and minimally invasive analytical tool with potential for the diagnosis of diseases. The potential for diagnostics can be improved by data post-processing. Hence, this study aims to evaluate the performance of preprocessing steps and multivariate analysis methods for the classification of normal tissues and pathological oral lesion spectra. A total of 80 spectra acquired from normal and abnormal tissues using optical fiber Raman-based spectroscopy (OFRS) were subjected to PCA preprocessing in the z-scored data set, and the KNN (K-nearest neighbors), J48 (unpruned C4.5 decision tree), RBF (radial basis function), RF (random forest), and MLP (multilayer perceptron) classifiers at WEKA software (Waikato environment for knowledge analysis), after area normalization or maximum intensity normalization. Our results suggest the best classification was achieved by using maximum intensity normalization followed by MLP. Based on these results, software for automated analysis can be generated and validated using larger data sets. This would aid quick comprehension of spectroscopic data and easy diagnosis by medical practitioners in clinical settings. (C) 2017 Optical Society of America. (AU)

• FAPESP's process: 14/16154-0 - Fluorescence spectroscopy for diagnosis of clinically similar skin lesions
• Grantee: Marcelo Saito Nogueira
• Support Opportunities: Scholarships in Brazil - Master
• FAPESP's process: 14/05978-1 - In vivo Raman spectroscopy for pathological processes of oral mucosa diagnosis
• Grantee: Luis Felipe das Chagas e Silva de Carvalho
• Support Opportunities: Scholarships in Brazil - Post-Doctoral