Premium
Development and validation of a neural network model for predicting growth of Salmonella Newport on diced Roma tomatoes during simulated salad preparation and serving: extrapolation to other serotypes
Author(s) -
Oscar Thomas P.
Publication year - 2018
Publication title -
international journal of food science and technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.831
H-Index - 96
eISSN - 1365-2621
pISSN - 0950-5423
DOI - 10.1111/ijfs.13767
Subject(s) - extrapolation , salmonella , serotype , artificial neural network , interpolation (computer graphics) , mathematics , biology , statistics , computer science , microbiology and biotechnology , artificial intelligence , bacteria , motion (physics) , genetics
Summary A study was undertaken to model growth of Salmonella on tomatoes for developing and validating a predictive model for use in risk assessment. Cylindrical portions (0.14 g) of Roma tomato pulp were inoculated with a low dose (0.89 log MPN ) of Salmonella Newport. The inoculated tomato portions were incubated for 0–8 h at 16–40 °C in 2 °C increments to obtain most probable number ( MPN ) data for model development and validation. A multiple‐layer feedforward neural network model with two hidden layers of two nodes each was developed. The proportion of residuals in an acceptable prediction zone ( pAPZ ) from −1 (fail‐safe) to 0.5 log (fail‐dangerous) was 0.93 (194/209) for dependent data and 0.96 (86/90) for independent data for interpolation. A pAPZ ≥0.7 indicated that the model provided acceptable predictions. Thus, the model was successfully validated. It was also validated for extrapolation to seven other Salmonella serotypes.