Abstract:

Four species of common food-borne pathogenic bacteria including Staphylococcus aureus, Escherichia coli,
Streptococcus fecal and Listeria monocytogenes were detected using a metal oxide sensor-based portable electronic nose
(PEN3) according to its different response to bacterial metabolites. Various fingerprints were established for the four
bacteria cultured for 2, 4, 6, 8, and 10 h as well as diluted 103, 105 and 107 times, respectively. The responses of PEN3 to
bacteria under different conditions of incubation time and concentration were analyzed using cluster analysis, principal
component analysis (PCA) and linear discriminant analysis (LDA). Results showed that volatile bacterial metabolites at
different incubation times could be well distinguished by PCA and LDA with the smallest inter- to intra-group variation
ratio. Meanwhile, a significant distinction among these four bacteria was observed even at lower concentration. These results
suggest that chemical sensor technology is feasible in detecting pathogenic bacteria at low concentrations.

Key words: PEN3 electronic nose, food-borne pathogenic bacteria, principal component analysis, linear discriminant analysis