Neural network model for growth of Salmonella Typhimurium in brain heart infusion broth

Abstract

Summary Models that predict growth of Salmonella as a function of variables in the current and previous environment are valuable tools for assessing the safety of food. Therefore, this study was undertaken to develop a model for growth of Salmonella Typhimurium in brain heart infusion broth as a function of previous pH (5.7–8.6), temperature (15–40 °C), pH (5.2–7.4) and time. Viable count data (log CFU mL−1) were modelled using a neural network approach. The variable impacts were 2.4% for previous pH, 29.0% for temperature, 4.9% for pH and 63.7% for time. The proportion of residuals in an acceptable prediction zone (pAPZ) from ‐1 (fail‐safe) to 0.5 log CFU mL−1 (fail‐dangerous) was 0.965 (1061/1100) for dependent data and 0.939 (386/411) for independent data for interpolation. A pAPZ ≥ 0.7 indicated that the model provided predictions with acceptable accuracy and bias. Thus, the model was successfully validated.

Use of ComBase data to develop an artificial neural network model for nonthermal inactivation of Campylobacter jejuni in milk and beef and evaluation of model performance and data completeness using the acceptable prediction zones method

“…In addition, it was based on the convention used in Com-Base (Baranyi & Tamplin, 2004). Based on previous studies (Oscar, 2009(Oscar, , 2014(Oscar, , 2017a(Oscar, , 2017b(Oscar, , 2018a(Oscar, , 2018b(Oscar, , 2021, data were tagged for testing as follows: (a) for a time course of 30 days at 9 and 21 days; (b) for a time course of 14 days at 4, 9, and 13 days; and (c) for a time course of 3 days at 1 and 2.25 days. This resulted in 270 (82%) data for training and 58 (18%) data for testing the ANN during training.…”

“…In the past, ANN was difficult to use in predictive microbiology applications. However, with the arrival of commercial software programs, it is now easy to develop ANN models for foodborne pathogens (Oscar, 2009(Oscar, , 2014(Oscar, , 2017a(Oscar, , 2017b(Oscar, , 2018a(Oscar, , 2018b(Oscar, , 2021.…”

“…Finally, over-training can be avoided by following the criteria of the APZ method when developing and validating ANN (Oscar, 2017b(Oscar, , 2018a(Oscar, , 2018b(Oscar, , 2020a(Oscar, , 2021. Thus, there is no reason not to use ANN in predictive microbiology applications.…”

Predictive model for growth of Salmonella Newport on Romaine lettuce

“…Five microliters of a frozen (−80 C), thawed, and resuspended stock culture of Salmonella Newport was added to 0.7 ml of BPW in a 1.5 ml polystyrene, micro-centrifuge tube. Stationary phase cells, as determined by a predictive model (Oscar, 2018b), were obtained by incubating the culture for 96 hr at 22 C.…”

Membrane fatty acids composition and fluidity modification in Salmonella Typhimurium by culture temperature and resistance under pulsed electric fields

“…Despite the fact that a large number of studies on the electroporation of cytomembrane have been reported (Saldaña et al ., ; Liu et al ., ), the mechanism of PEF treatment to inactivation micro‐organism is still not fully understood. In addition, when under external stress the cytomembrane makes alterations in its fatty acid composition, which affects many functions of the cell, especially membrane fluidity (Oscar, ; Niu et al ., ), there appears to be a relationship between membrane fluidity and bacterial resistance to environmental stresses (Niu et al ., ). In general, lower pH, greater heat and pressure resistance of bacteria are found in cells with a lower membrane fluidity (Cebrián et al ., ; Ou et al ., ), which might provide a new clue to explain the inactivation of micro‐organisms by PEF treatment.…”