C.H Herremans

Application of artificial neural networks as a non-linear modular modeling technique to describe bacterial growth in chilled food products

In many chilled, prepared food products, the effects of temperature, pH and %NaCl on microbial activity interact and this should be taken into account. A grey box model for prediction of microbial growth is developed. The time dependence is modeled by a Gompertz model-based, non-linear differential equation. The influence of temperature, pH and %NaCl reflected in the model parameters is described by using low-complexity, black box artificial neural networks (ANNs). The use of this non-linear modeling technique makes it possible to describe more accurately interacting effects of environmental factors when compared with classical predictive microbiology models. When experimental results on the influence of other environmental factors become available, the ANN models can be extended simply by adding more neurons and/or layers.

Kinetics for Isobaric−Isothermal Inactivation of Bacillus subtilis α-Amylase

Isobaric−isothermal inactivation of Bacillus subtilis α‐amylase (BSA, 15 mg/mL in Tris‐HCl at pH 8.6) in the pressure range 1–750 MPa and the temperature range 25–85 °C could be accurately described by a first‐order kinetic model. The kinetic parameters (k, Ea, and Va) were calculated at different pressure and temperature levels. At reference temperature (40 °C) and reference pressure (500 MPa), isobaric−isothermal inactivation was characterized by an Ea value of 74.8 kJ/mol, a Va value of −23.6 cm3/mol, and an inactivation rate constant of 0.0343 min−1. The influence of 15% glycerol on thermal and pressure−temperature stability of BSA was investigated. In both cases, a stabilizing effect of this additive was found, since the kref value was significantly reduced. Furthermore, a pressure−temperature kinetic diagram, indicating the possible synergistic and antagonistic effects of pressure and temperature on the inactivation of BSA, was constructed. Based on this diagram, a model describing the dependence of the inactivation rate constant on pressure and temperature, in the pressure range 250–750 MPa, was formulated.

Modeling the kinetics of isobaric-isothermal inactivation of Bacillus subtilis α-amylase with artificial neural networks

Abstract During the isobaric-isothermal Inactivation of the enzyme α-amylase a simple first order inactivation kinetic is observed. However, the temperature and pressure dependence of the inactivation rate constant k (min −1 ) is more complex. A non-synergetic (cumulative) model inspired by the Arrhenius law is proposed as a non-linear description. Significant model deficiency is observed at non-intermediate values of pressure and temperature. Due to the lack of sufficient knowledge of the underlying biological and physical mechanisms, in this paper a black box modeling approach is made using artificial neural networks. The resulting artificial neural network structure is able to predict the combined effect of pressure and temperature on the inactivation rate constant k (min −1 ) of α-amylase without significant increase of the model complexity as compared to the Arrhenius type model. The accuracy of the ANN model parameters is evaluated using the concept of joint confidence regions.

Evaluation of model parameter accuracy by using joint confidence regions: application to low complexity neural networks to describe enzyme inactivation

An existing low complexity, black box artificial neural network model (ANN model) is investigated towards its more general applicability in the field of high isobaric–isothermal inactivation of enzymes. The use of this non-linear modeling technique makes it possible to describe accurately synergistic effects of pressure and temperature in contrast with more classical models used in this novel area of food processing.

Modeling the Kinetics of Isobaric-Isothermal Inactivation of Bacillus Subtilis α-Amylase with Artifical Neural Networks

Abstract During the isobaric-isothermal inactivation of the enzyme α-amylase a simple first order inactivation kinetic is observed. However, the pressure and temperature dependence of the inactivation rate constant k [1/min] is more complex. A non synergetic (cumulative) model inspired by the Arrhenius law is proposed. Significant model deficiency is observed at nonintermediate values of pressure and temperature. Due to the lack of sufficient knowledge of the underlying biological mechanisms, a black box modeling approach is made using Artificial Neural Networks . Several network structures are evaluated. The resulting artificial neural network structure relaxes the model deficiency at non-intermediate pressure and temperature values, without increase of the model complexity as compared to the Arrhenius type model. As a result, the model is able to describe the combinatory effect of pressure and temperature on the inactivation rate constant of α-amylase.