Laser induced breakdown spectroscopy for the discrimination of Candida strains

Abstract

The present study reports the evaluation of Laser Induced Breakdown Spectroscopy (LIBS) and Neural Networks (NN) for the discrimination of different strains of various species of Candida. This genus of yeast was selected due to its medical relevance as it is commonly found in cases of fungal infection in humans. Twenty one strains belonging to seven species of Candida were included in the study. Scanning Electron Microscopy with Energy-Dispersive X-ray Spectroscopy (SEM-EDS) was employed as a complementary technique to provide information about elemental composition of Candida cells. The use of LIBS spectra in combination with optimized NN models provided reliable discrimination among the distinct Candida strains with a high spectral correlation index for the samples analyzed, without any false positive or false negative. Therefore, this study indicates that LIBS-NN based methodology has the potential to be used as fast fungal identification or even diagnostic method.

Keywords: Candida; Candidiasis; Laser Induced Breakdown Spectroscopy; Neural Networks.