Electronic nose classification and differentiation of bacterial foodborne pathogens based on support vector machine optimized with particle swarm optimization algorithm

Abstract Rapid detection of bacterial foodborne pathogens is crucial in reducing the incidence of diseases associated with food contaminated with pathogens and toxins. This article presents a classification model of support vector machine (SVM) together with a metaheuristic framework using genetic algorithm (GA), particle swarm optimization (PSO), and grid searching (GS) optimization algorithms for bacterial foodborne pathogen classification and differentiation. LDA and SVM showed classification accuracies of training (90.3% and 91%) and prediction (89.5% and 90.6%), respectively using the Matlab classification learner app. Optimization of the modeling parameters c and g for SVM were performed to increase efficiency and classification accuracy. Simulated results show classification accuracies of 100% (training set) and 98.95% (prediction set) for five selected bacterial pathogens acquired using electronic nose dataset and PSO‐SVM model. Electronic nose recognition system made up of 12 metal oxide semiconductor sensors produced different distinctive response signals for each bacterial and could differentiate Escherichia coli , Escherichia coli O157: H7, Listeria monocytogenes , Salmonella enteritidis , and Salmonella Typhimurium . PSO‐SVM algorithm can be efficiently used in bacterial discrimination at the species and strain level by electronic nose and has the good ability both in learning and in generalization. Practical Applications Existing microbial methods depend on traditional culture‐based methods which are time‐wise lengthy, require trained and qualified personnel, and are not suitable as point‐of‐use (POU) sensing devices. The application of electronic nose for the detection and classification of bacterial foodborne pathogens have shown great promise and proven to be effective with increasing sensitivity and selectivity as compared to traditional methods. The ability for E‐nose to discriminate between individual bacteria colonies at both species and strain level is of great public health importance since most bacteria have many strains. Virulence and pathogenicity are often associated with only a subset of these strains and it is essential for a method to be able to differentiate between pathogenic and nonpathogenic strains during a foodborne outbreak.