Evaluation of a new mathematical model to describe Clostridium perfringens growth during the cooling of cooked ground beef

Application of deep neural techniques in predictive modelling for the estimation of Escherichia coli growth rate

AIMS

To develop a predictive model for Escherichia coli using deep neural networks.

METHODS AND RESULTS

Batch experiments are conducted at different temperatures closer to optimum value (36·5°C, 37°C, 37·5°C, 38°C and 38·5°C) to obtain the growth curves of E .coli K-12. Two primary models namely modified Gompertz and new logistic are chosen. Three secondary models namely Gaussian, nonlinear autoregressive eXogenous (NARX) model and long short-term memory (LSTM) are developed. The novelty in this paper is the development of secondary models using artificial neural network (ANN) and deep network. The performance measures chosen to compare the developed primary and secondary models are correlation coefficient (R² ), root-mean-square error (RMSE) and accuracy factor (A_f ). Results show that modified Gompertz model has better R² (0·99) and RMSE (0·019) when compared to new logistic model. Also, the deep network model outperforms other secondary models. Based on the primary and novel secondary model, a predictive model (tertiary model) is developed with improved accuracy and is validated.

CONCLUSIONS

The proposed predictive model exhibit good validation results in terms of RMSE and R² values and can be applied for determining the growth rate of E. coli at a particular temperature value.

SIGNIFICANCE AND IMPACT OF THE STUDY

The proposed model can be used in food processing industries during enzyme production such as Chymosin, to predict the growth rate of E. coli as a function of temperature. Also, the developed LSTM and NARX models can be used to predict maximum specific growth rate of other microbial strains with proper training.

Evaluating the Performance of a New Model for Predicting the Growth of Clostridium perfringens in Cooked, Uncured Meat and Poultry Products under Isothermal, Heating, and Dynamically Cooling Conditions

Clostridium perfringens type A is a significant public health threat and its spores may germinate, outgrow, and multiply during cooling of cooked meats. This study applies a new C. perfringens growth model in the USDA Integrated Pathogen Modeling Program-Dynamic Prediction (IPMP Dynamic Prediction) Dynamic Prediction to predict the growth from spores of C. perfringens in cooked uncured meat and poultry products using isothermal, dynamic heating, and cooling data reported in the literature. The residual errors of predictions (observation-prediction) are analyzed, and the root-mean-square error (RMSE) calculated. For isothermal and heating profiles, each data point in growth curves is compared. The mean residual errors (MRE) of predictions range from -0.40 to 0.02 Log colony forming units (CFU)/g, with a RMSE of approximately 0.6 Log CFU/g. For cooling, the end point predictions are conservative in nature, with an MRE of -1.16 Log CFU/g for single-rate cooling and -0.66 Log CFU/g for dual-rate cooling. The RMSE is between 0.6 and 0.7 Log CFU/g. Compared with other models reported in the literature, this model makes more accurate and fail-safe predictions. For cooling, the percentage for accurate and fail-safe predictions is between 97.6% and 100%. Under criterion 1, the percentage of accurate predictions is 47.5% for single-rate cooling and 66.7% for dual-rate cooling, while the fail-dangerous predictions are between 0% and 2.4%. This study demonstrates that IPMP Dynamic Prediction can be used by food processors and regulatory agencies as a tool to predict the growth of C. perfringens in uncured cooked meats and evaluate the safety of cooked or heat-treated uncured meat and poultry products exposed to cooling deviations or to develop customized cooling schedules. This study also demonstrates the need for more accurate data collection during cooling.