Author
|
Submitted to: Journal of Food Protection
Publication Type: Peer Reviewed Journal Publication Acceptance Date: 6/3/2015 Publication Date: 10/1/2015 Citation: Oscar, T.P. 2015. Neural network model for survival and growth of Salmonella 8,20:-:z6 in ground chicken thigh meat during cold storage: extrapolation to other serotypes. Journal of Food Protection. 78(10):1819-1827. doi: 10.4315/0362-028X.JFP-15-093. Interpretive Summary: Time of cold storage of chicken meat in the home refrigerator ranges from 0 to 5 days with a most likely time of 2 days whereas the temperature of cold storage in the home refrigerator can range from 18F to 60F with a most likely temperature of 40F. Results of this study indicated that as long as consumers keep chicken meat at less than 46F, the number of Salmonella, if present, will stay the same during 8 days of cold storage. However, the number of Salmonella will increase from 50 to 500 cells during 2 days of storage at 54F whereas it will increase from 50 to 500,000 during storage for 2 days at 60F. A model that predicts the number of Salmonella in chicken meat as a function of time and temperature of cold storage was developed. The new model will be a valuable tool for helping chicken producers better assess and manage the impact of consumer cold storage conditions and practices on the risk of salmonellosis from chicken obtained at retail. Technical Abstract: Mathematical models that predict behavior of human bacterial pathogens in food are valuable tools for assessing and managing this risk to public health. A study was undertaken to develop a model for predicting behavior of Salmonella 8,20:-:z6 in chicken meat during cold storage and to determine how well the model predicts behavior of other serotypes of Salmonella stored under the same conditions. For model development, ground chicken thigh meat (0.75 cm3) was inoculated with 1.7 log of Salmonella 8,20:-:z6 followed by storage for 0 to 8 days at -8 to 16 degrees C. An automated mini-most probable number method was used for enumeration of Salmonella. Commercial software (Excel + NeuralTools) were used to develop a multi-layer feedforward neural network model with one hidden layer of two nodes. Model performance was evaluated using the acceptable prediction zone method. Number of Salmonella in ground chicken thigh meat stayed the same (P is greater than 0.05) during 8 days of storage at -8 to 8 degrees C but increased (P less than 0.05) over time of storage at 9 to 16 degrees C. The proportion of residuals (observed – predicted) in an acceptable prediction zone (pAPZ) from -1 log (fail-safe) to 0.5 log (fail-dangerous) was 0.939 for data (n = 426) used in model development. The model had pAPZ of 0.944 or 0.954 when it was extrapolated to test data (n = 108/serotype) for other serotypes (Typhimurium var 5-, Kentucky, Typhimurium, Thompson) of Salmonella in ground chicken thigh meat stored for 0 to 8 days at -4, 4, 12, or 16 degrees C under that same experimental conditions. A pAPZ is greater than or equal to 0.7 indicates that a model provides predictions with acceptable bias and accuracy. Thus, results indicated that the model provided valid predictions of survival and growth of Salmonella 8,20:-:z6 in ground chicken thigh meat stored for 0 to 8 days at -8 to 16 degrees C and that the model was validated for extrapolation to four other serotypes of Salmonella. |
