Multivariate analysis of laryngeal fluorescence spectra recorded in vivo

Background and Objective: The potential of using various multivariate analysis methods for classification of fluorescence spectra acquired in vivo from laryngeal tissues in Patients was investigated. Study Design/Materials and Methods: Autofluorescence spectra were measured on 29 normal tissue sites and 25 laryngeal lesions using 337-nm excitation. Four different multivariate analysis schemes were applied. Laryngeal fluorescence spectra from patients who had been administered δ-aminolevulinic acid (ALA) were obtained using 405-nm excitation and were classified using partial least squares discriminant analysis (PLS-DA). Results: For autofluorescence spectra, logistic regression based on principal component analysis (PCA) or PLS, or PLS-DA all resulted in sensitivities and specificities around 90% for lesion vs. normal. Using ALA and 405-nm excitation gave a sensitivity of 100% and a specificity of 69%. Conclusion: Multivariate analysis of fluorescence spectra could allow classification of laryngeal lesions in vivo with high sensitivity and specificity. PLS performs at least as well as PCA, and PLS-DA performs as well as logistic regression techniques on these data.