Rapid Identification and Characterization of Pathogenic Bacteria Isolates Using Fluorescence Excitation Spectroscopy and Spectral Deconvolution

Bacterial detection and identification are crucial tasks in today's world due to increasing casualties and severity caused by bacteria in hospital‐acquired infections, food‐borne infections, etc. This warrants an urgent need for rapid detection and identification of different bacterial strains. Fluorescence spectroscopy is a widely used technique for bacteria characterization owing to its high sensitivity, real‐time, and simple measurement capabilities. In this work, fluorescence excitation spectra of 200 bacteria isolates belonging to eight different species were acquired under identical conditions with 25 isolates per species. The fluorescence excitation spectra were subjected to spectral deconvolution to extract four Gaussian components viz., tryptophan in protein under buried and exposed conditions, nucleic acids and anthranilic acid. Linear discriminant analysis applied to normalized fluorescence excitation spectra of 200 samples classified the bacterial strains with 100% accuracy using leave‐one‐out cross‐validation.