Growth Characteristics and Development of a Predictive Model for Escherichia coli in Rice Cakes

Introduction: Rice cake is a suitable food for microbial growth due to its high A_w, neutral pH, and storage temperature. However, a little study on microbiological safety of rice cakes has been performed.

Purpose: The objective of this study was to develop a predictive models for the growth of Escherichia coli in three kinds of rice cakes (Garaetteok, Sirutteok, and Gyeongdan) as a function of temperature (10-40 °C)

Methods: The growth data was fitted into the modified Gompertz model as primary models. The polynomial quadratic model was selected as secondary models and used the data such as specific growth rate (SGR), and lag time (LT) obtained from the primary models. The performances of predictive modeling were evaluated using mean square error (MSE), accuracy factor (A_f), bias factor (B_f) and coefficient of determination (r²). 

Results: The primary model of E. coli showed a good fit to Gompertz equation to obtain SGR and LT of each rice cakes. It was found that the values of SGR tends to increase in proportion to temperature and the value of LT decreased in inverse proportion to temperature in all three rice cakes. The appropriateness of secondary polynomial model was verified by a set of MSE, A_f, B_f and r². The MSE values of SGR and LT models in Garaetteok are 0.0006 and 0.28158, and r² are 0.9864 and 0.9962. The A_f values were 1.1628 and 1.3548, and B_f are 0.9484 and 0.9422, respectively. The MSE values of SGR and LT models in Sirutteok were 0.0008 and 0.0005 and r² are 0.9999 and 0.9999, respectively. The A_f values were 1.0575 and 1.0021, and B_f are 1.0575 and 0.9979. Lastly, the MSE values of SGR and LT models in Gyeongdan were both 0.000 and r² are both 1.000. The A_f values of SGR and LT models in Gyeongdan were 1.022 and 1.000, and B_f are 1.022 and 1.000, respectively.

Significance: These predictive models could be used for quantitative microbial risk assessment of E. coli in rice cakes.