The Identification of Listeria Monocytogenes Based on the Electronic Nose

摘要

Our previous work on electronic nose can effectively detect several major foodborne pathogens,which is based on principle component analysis (PCA) and cluster analysis (CA) method. These building methods cannot identify two strains of Listeria monocytogenes,which have the serotypes of 4b and 4c respectively. To resolve this problem,we propose a neural network method which can differentiate these two strains. The specifically constructed neurons can get feature data from the E-nose output text files. The first layer will detect the volatile metabolites of 4 species of Listeria spp. and 9 strains of L. monocytogenes. This is because the volatile metabolites in the culture medium of 4 species of Listeria spp. could be well distinguished by PCA. The second layer use selected differential feature data to identify 4b and 4c serotypes. Experiment shows that the improved E-nose method has good stability and repeatability. This study indicates the odor fingerprint based on detecting microbial volatile metabolites can be enhanced with new feature extracted and used in a lot of pathogen identification.