Predictive modeling in food mycology using adaptive neuro-fuzzy systems

Fungal growth leads to spoilage of food and animal feeds and to formation of mycotoxins and potentially allergenic spores. There is a growing interest in predictive modeling microbial growth as an alternative to time consuming traditional, microbiological enumeration techniques. Several statistical models have been accounted to describe the growth of different micro-organisms. However neural networks, as highly nonlinear approximator scheme, have the potential of modeling some complex, phenomena better than the others. The application of adaptive neuro-fuzzy systems in predictive microbiology is presented in this paper. This technique is used to build up a model of the joint effect of water-activity, pH level and temperature to predict the maximum specific growth rate of the Ascomycetous Fungus Monascus Ruber. The proposed scheme is compared against standard neural network approaches. Neuro-fuzzy systems offer an alternative and powerful technique to model microbial kinetic parameters and could thus become an efficient tool in predictive mycology.