Detection and identification of selected bacteria, inoculated on chicken breast, using near infrared spectroscopy and chemometrics

Near-infrared transflectance spectra of chicken breast muscle inoculated with Listeria innocua FH 2333, Pseudomonas fluorescens 2088, Pseudomonas putida 49128, Pseudomonas mendocina or Escherichia coli K12 were investigated in order to examine the potential of NIRS to detect microbes on a food matrix. Inoculated samples were stored at 36 °C for 24 h. Two control groups were used in this study, one of fresh uninoculated samples analysed directly after removal of the muscle from the carcass and another of uninoculated chicken stored at 36 °C for 24 h. The findings of this study suggest the NIR spectroscopy can detect and differentiate between inoculated and uninoculated chicken breast muscle. In addition, it can discriminate fresh from non-fresh (in this case samples that are inoculated or not and stored at 36 °C) samples. On the other hand, NIR spectroscopy failed to differentiate between the five different strains used as inoculates for this study. These results suggest that the make up of the microflora of chicken fillets is not the deciding factor in detecting and differentiating chicken fillet samples. Instead, it seems that the total quantity of microflora present in conjunction with the physical changes associated with spoilage and, in particular, the differences in microbial concentration and level of muscle decomposition of inoculated and uninoculated samples, may be responsible for the separations observed in this analysis. This reason could also be the deciding factor for the separations achieved between the fresh samples and the other two groups.