García-Gimeno, R.M., *, Rodríguez-Pérez, M.R., Hervás-Martínez, C. and Zurera-Cosano, G.,
Modelling the growth of Leuconostoc mesenteroides by means artificial neural network, International Journal of Food Microbiology, 2004

ABSTRACT

The combined effects of different temperatures (10.5 to 24.5ºC), pH level (5.5 to 7.5), sodium chloride levels (0.25 to 6.25%) and sodium nitrite levels (0 to 200 ppm) on the predicted growth rate, lag-time and yEnd of Leuconostoc mesenteroides under aerobic and anaerobic conditions was studied using an artificial neural network-based model (ANN) in comparison with response surface methodology (RS). The individual ANN model developed provided reliable estimates of the three kinetic parameters studied, with SEP values ranged between 2.82% and 14.60%, betters than those of RS model (17.62-27.63%). ANN predictive growth models are a valuable tool, enabling swift determination of L. mesenteroides growth rate lag-time and y End growth parameters.

Keywords: predictive microbiology; Leuconostoc mesenteroides; artificial neural network; response surface model

INTRODUCTION

Growth predictive models are currently accepted as informative tools that assist in rapid and cost-effective assessment of microbial growth for product development, risk assesment and education purposes (Ross, 1999). Although during the past few years, much effort has been directed to developing models describing the combined effects of environmental factors on microbial growth of pathogens in foods (Ross et al., 2000; Devlieghere et al, 2001; García-Gimeno et al., 2003; Zurera-Cosano et al., 2004), predictive microbiology has been used to forecast the growth of spoilage microorganisms in order to study the shelf life of a food product. Specific spoilage organisms are selected for certain food products and used as test organisms as Brochothrix thermosphacta (Baranyi et al., 1995), Pseudomonas (Neumeyer et al., 1997a), Lactobacillus sake (Devlieghere et al., 1998), Lactobacillus curvatus (Wijtzes et al., 2001), o Lactobacillus plantarum (García-Gimeno et al., 2002).

Although Response Surface Methodology are the most frequent techniques used to describe the relationships between the combination of factors and the growth curve parameters (Devlieghere et al., 1998; Zurera-Cosano et al 2004), more recently, a number of new models have been introduced, some involving the application of Artificial Neural Networks (ANN), which can estimate kinetic parameters from the growth curves and also from factors affecting microbial growth (Hajmeer et al., 1997; Geeraerd et al., 1998; Hervás et al., 2001; Jeyamkondan et al., 2001; Lou and Nakai, 2001; García-Gimeno et al., 2002).

Once the best model has been obtained by the training process, this is tested with a data set, obtained in a similar way and denominated “generalization data set”, to evaluate the prediction capacity of the proposed model. This could be equivalent to what has been called by other predictive microbiology authors “mathematical validation” (Van Impe et al., 1998). The generalization capacity is evaluated by means of the Standard Error of Prediction percentage (% SEP), which is a relative typical deviation of the mean prediction values and has the advantage, compared to other error measures, that it is not dependent on the magnitude of the measurements.

The ability of the ANN technique in modeling the growth of Leuconostoc mesenteroides under different experimental conditions of temperature, pH, salt and nitrite concentrations in aerobic and anaerobic conditions on the growth rate, lag-time and maximum population density in comparison of Response Surface Methodology (RSM) was studied.

REFERENCES

Arinder, P. and Borch, E. 1999. Validation of mathematical models for predicting growth of Pseudomonas spp. Predictive microbiology applied to chilled food preservation. In: Refrigeration Science and Technology Proceedings, pp. 185-193.

Baranyi, J. and Roberts, T.A. 1994. A dynamic approach to predicting bacterial growth in food. Int. J. Food Microbiol., 23, 277-294.

Baranyi, T., Robinson, T.P., Kaloti, A. and Mackey, B.M. 1995. Predicting growth of Brochothrix thermosphacta at changing temperature. Int. J. Food Microbiol., 27, 61-75.

Basheer, I. and Hajmeer, M. 2000. Artificial neural networks: fundamentals, computing, design and application. J. Microbiological Methods, 43, 3-31.

Bharathi, S., Ramesh, M.N. and Varadaraj, M.C. 2001. Predicting the behavioural pattern of Escherichia coli in minimally processed vegetables. Food Control, 12, 275-284.

Dalgaard, P. 2000. Fresh and lightly preserved seafood. In: Shelf life evaluation of foods. 2nd Edition Man, C. M. D. and Jones, A. A. (Eds.). Aspen Publishing Inc. Maryland, pp. 110- 139.

Devlieghere, F., Debevere, J. and Van Impe, J. 1998. Effect of dissolved carbon dioxide and temperature on the growth of Lactobacillus sake in modified atmospheres. Int. J. Food Microbiol., 41, 231-238.

Devlieghere, F., Geeraerd, A.H., Versyck, K.J., Van De Waetere, B., Van Impe, J. and Debevere, J. 2001. Growth of Listeria monocytogenes in modified atmosphere packed cooked meat products: a predictive model. Food Microbiol., 18, 53-66.

García-Gimeno, R.Mª, Hervás-Martínez, C. and de Silóniz, MªI. 2002. Improving artificial neural networks with a pruning methodology and genetic algorithms for their application in microbial growth prediction in food. Int. J. Food Microbiol., 72, 19-30.

García- Gimeno, R.M., Hervás- Martínez, C., Barco- Alcalá, E., Zurera- Cosano, G. and Sanz- Tapia, E. 2003. An artificial neural network approach to Escherichia coli O157:H7 growth estimation. J. Food Sci., 68 (1), 1-7.

Geeraerd, A., Herremans, C., Cenens, C. and Van Impe J. 1998. Applications of artificial neural networks as a non-linear modular modeling technique to describe bacterial growth in chilled food products. Int. J. Food Microbiol., 44, 49-68.

Hajmeer, M., Basheer, I. and Najjar, Y. 1997. Computational neural networks for predictive microbiology II. Application to microbial growth. Int. J. Food Microbiol., 34, 51-66.

Hassibi, B. and Storck, D. 1993. Second-order derivatives for network pruning: optimal brain surgeon. En: Advances in Neural Information Processing (NIPS-5) vol. 5. S. Hanson, S. Cowan and C.L. Giles, Editors. Morgan Kaufmann, San Mateo, CA, pp. 164–171.

Fecha de impresión 05/07/2004, RN vs RS 1.doc 17 Hervás, C., Zurera, G., García, R.M. and Martínez, J. 2001. Optimization of computational neural network for its application to the prediction of microbial growth in foods. Food Sci. Technol. Int., 7, 159-163.

Huis in't Veld, J. H. J. 1996. Microbial and biochemical spoilage of foods: an overview. Int. J. Food Microbiol., 33, 1-18.

Jeyamkondan, S., Jayas, D. and Holley, R. 2001. Microbial growth modeling with artificial neural networks. Int. J. Food Microbiol., 64, 343-354.

Juneja, V.K., Eblen, B.S. and Marks, H.M. 2001. Modeling non-linear survival curves to calculate thermal inactivation of Salmonella in poultry of different fat levels. Int. J. Food Microbiol., 70, 37-51.

Krogh, A. and Hertz, J. 1992. A simple weigth decay can improve generalization. In: Touretzky, D.S. (Ed.), Advances in Neural Information- Processing Systems, vol. 4. Morgan Kaufmann, San Mateo, CA, pp. 950-957.

Le Cun, Y., Denker, J. and Solla, S. 1990. Optimal brain damage. En: Advances in neural information- processing systems 2. D. S. Touretzky (Ed). San Mateo, CA: Morgan Kaufmann, pp. 598-605.

Lebert, I., Robles-Olvera, V. and Lebert, A. 2000. Application of polynominal models to predict growth of mixed cultures of Pseudomonas spp. and Listeria in meat. Int. J. Food Microbiol., 61, 27-39.

Lou, W. and Nakai, S. 2001. Application of artificial neural networks for predicting the thermal inactivation of bacteria: a combined effect of temperature, ph and water activity. Food Research Int., 34, 573-579.

McCann, T.L., Eifert, J.D., Gennings, C., Schilling M.W. and Carter, Jr. W.H. 2003. A predictive model with repeated measures analysis of Staphylococcus aureus growth data. Food Microbiol., 20, 139-147.

Mittal, G. and Zhang, J. 2000. Prediction of freezing time for food products using a neural network. Food Research Int., 33, 557-562.

Najjar, Y., Basheer, I. and Hajjmer, M. 1997. Computational neural networks for predictive microbiology: I. Methodology. Int. J. Food Microbiol., 34, 27-49.

Nauta, M.J., Litman , S., Barker, G.C. and Carlin, F. 2003. A retail and consumer phase model for exposure assessment of Bacillus cereus. Int. J. Food Microbiol., 83, 205-218.

Neumeyer, K., Ross, T. and McMeekin, T.A. 1997a. Development of a predictive model to describe the effects of temperature and water activity on the growth of spoilage Pseudomonads. Int. Food Microbiol., 38, 45-54. Fecha de impresión 05/07/2004, RN vs RS 1.doc 18

Neumeyer, K., Ross, T. Thomson, G. and McMeekin, T.A. 1997b. Validation of a model describing the effects of temperature and water activity on the growth of psychrotrophic pseudomonads.Int. Food Microbiol., 38, 55-63.

Robinson, T.P., Ocio, M.J. Kaloti, A. and Mackey, B.M. 1998. The effect of the growth environment on the lag phase of Listeria monocytogenes. Int. J. Food Microbiol., 44, 83-92.

Ross, T. 1999. Predictive food microbiology models in the meat industry. Meat and Livestock Australia, Sydney, Australia, pp. 196.

Ross, T., Dalgaard, P. and Tienungoon, S. 2000. Predictive modelling of the growth and survival of Listeria in fishery products. Int. J. Food Microbiol., 62, 231-245.

Valik, L. & Pieckova, E. 2001. Growth modelling of heat-resistant fungi : the effect of water activity. International Journal of Food Microbiology , 63, 11-17.

Van Impe, J.F., Versyck, K.J. and Geeraerd, A.H. 1996. Validation of predictive models: definitions and concepts. In: COST 914. Predictive modelling of microbial growth and survival in foods. Luxemburgo. European Comission, Brussels, pp. 31-38.

Wijtzes, T., Rombouts, F.M., Kant-Muermans, M.L.T., van’t Riet, K. and Zwietering, M.H. 2001. Development and validation of a combined temperature, water activity, pH model for bacterial growth rate of Lactobacillus curvatus. Int. J. Food Microbiol., 63, 57-64.

William, P.M. 1995. Bayesian regularization and pruning use a Laplace prior. Neural Comput., 7, 117-143.

Zhang, G. and Holley, R.A. 1999. Development and PFGE monitoring of dominance among spoilage lactic acid bacteria from cured meats. Food Microbiol., 16, 633-644.

Zurera-Cosano, G., Castillejo- Rodríguez, A.M., García- Gimeno, R.M. and Rincón- León, F. (2004). Response Surface vs Davey Model for the estimation of Staphylococcus aureus growth under different experimental conditions. J. Food Prot., 67 (5)