Predictive Food Microbiology Research Articles

Prevalence, Characteristics, and Selection of Bacillus cereus Subgroups from Dairy Products for Challenge Testing and Predictive Model Development

Prevalence, toxin gene profiles, lactose fermentation, and growth responses of B. cereus sensu lato subgroups in various dairy and dairy alternative products and ingredients were studied to identify relevant isolates for challenge testing and model development to predict and manage growth responses. Out of 71 examined products or ingredients, 51 B. cereus s.l. isolates were obtained from 35 positive samples (49% prevalence). These 51 isolates along with 18 additional dairy isolates and 12 B. cereus s.l. reference strains were identified using MALDI-TOF. The 81 isolates were further characterized by panC sequencing, testing for cold shock and toxin genes (cspA; hbl, nhe, CytK and ces), lactose fermentation, and study of growth rates (µmax) under various conditions (45 °C, 10 °C, 6% NaCl, pH 5.1), resulting in 298 µmax-values. These conditions were selected to differentiate mesophilic and psychrotolerant strains and to identify tolerant isolates. Dairy powders (83%), pasteurized upconcentrated cheese whey (43%), and cheeses (42%) had the highest prevalences of B. cereus s.l. and the highest concentrations in positive samples (5–100 CFU/g or ml). The panC groups II, III, IV, VI, and VIII were detected among the dairy isolates, with 97% harboring one or more toxin genes. Lactose fermentation was observed in 42% of isolates, with lactose-fermenting B. cereus s.l. isolates of panC groups III and IV dominant in dairy powders. Growth rates of B. cereus s.l. varied considerably, among and within panC groups of the dairy isolates. Based on the highest growth rates at 45 °C, 10 °C, 6% NaCl, pH 5.1, panC group membership, toxin genes profiles, and lactose fermentation ability, two cocktails of “mesophilic” or “psychrotolerant” isolates were selected. These strain cocktails can be used in future challenge testing and predictive food microbiology studies to evaluate and manage the growth of B. cereus s.l. in dairy products and ingredients.

The Use of Predictive Microbiology for the Prediction of the Shelf Life of Food Products.

Microbial shelf life refers to the duration of time during which a food product remains safe for consumption in terms of its microbiological quality. Predictive microbiology is a field of science that focuses on using mathematical models and computational techniques to predict the growth, survival, and behaviour of microorganisms in food and other environments. This approach allows researchers, food producers, and regulatory bodies to assess the potential risks associated with microbial contamination and spoilage, enabling informed decisions to be made regarding food safety, quality, and shelf life. Two-step and one-step modelling approaches are modelling techniques with primary and secondary models being used, while the machine learning approach does not require using primary and secondary models for describing the quantitative behaviour of microorganisms, leading to the spoilage of food products. This comprehensive review delves into the various modelling techniques that have found applications in predictive food microbiology for estimating the shelf life of food products. By examining the strengths, limitations, and implications of the different approaches, this review provides an invaluable resource for researchers and practitioners seeking to enhance the accuracy and reliability of microbial shelf life predictions. Ultimately, a deeper understanding of these techniques promises to advance the domain of predictive food microbiology, fostering improved food safety practices, reduced waste, and heightened consumer confidence.

Prediction of Pseudomonas spp. Population in Food Products and Culture Media Using Machine Learning-Based Regression Methods.

Machine learning approaches are alternative modelling techniques to traditional modelling equations used in predictive food microbiology and utilise algorithms to analyse large datasets that contain information about microbial growth or survival in various food matrices. These approaches leverage the power of algorithms to extract insights from the data and make predictions regarding the behaviour of microorganisms in different food environments. The objective of this study was to apply various machine learning-based regression methods, including support vector regression (SVR), Gaussian process regression (GPR), decision tree regression (DTR), and random forest regression (RFR), to estimate bacterial populations. In order to achieve this, a total of 5618 data points for Pseudomonas spp. present in food products (beef, pork, and poultry) and culture media were gathered from the ComBase database. The machine learning algorithms were applied to predict the growth or survival behaviour of Pseudomonas spp. in food products and culture media by considering predictor variables such as temperature, salt concentration, water activity, and acidity. The suitability of the algorithms was assessed using statistical measures such as coefficient of determination (R2), root mean square error (RMSE), bias factor (Bf), and accuracy (Af). Each of the regression algorithms showed appropriate estimation capabilities with R2 ranging from 0.886 to 0.913, RMSE from 0.724 to 0.899, Bf from 1.012 to 1.020, and Af from 1.086 to 1.101 for each food product and culture medium. Since the predictive capability of RFR was the best among the algorithms, externally collected data from the literature were used for RFR. The external validation process showed statistical indices of Bf ranging from 0.951 to 1.040 and Af ranging from 1.091 to 1.130, indicating that RFR can be used for predicting the survival and growth of microorganisms in food products. Therefore, machine learning approaches can be considered as an alternative to conventional modelling methods in predictive microbiology. However, it is important to highlight that the prediction power of the machine learning regression method directly depends on the dataset size, and it requires a large dataset to be employed for modelling. Therefore, the modelling work of this study can only be used for the prediction of Pseudomonas spp. in specific food products (beef, pork, and poultry) and culture medium with certain conditions where a large dataset is available.

Development of prediction software to describe total mesophilic bacteria in spinach using a machine learning-based regression approach.

The purpose of this study was to create a tool for predicting the growth of total mesophilic bacteria in spinach using machine learning-based regression models such as support vector regression, decision tree regression, and Gaussian process regression. The performance of these models was compared to traditionally used models (modified Gompertz, Baranyi, and Huang models) using statistical indices like the coefficient of determination (R2) and root mean square error (RMSE). The results showed that the machine learning-based regression models provided more accurate predictions with an R2 of at least 0.960 and an RMSE of at most 0.154, indicating that they can be used as an alternative to traditional approaches for predictive total mesophilic. Therefore, the developed software in this work has a significant potential to be used as an alternative simulation method to traditionally used approach in the predictive food microbiology field.

Development of a New Modelling Approach and Performance Evaluation of Meta-heuristic Optimization Algorithms for the Prediction of Kinetic Growth Parameters for Pseudomonas spp. in Fish

The main aim of the current work was to build up a new mathematical modelling approach in predictive food microbiology field for the prediction of growth kinetics of microorganisms. For this purpose, the bacterial growth data of Pseudomonas spp. in whole fish (gilt-head seabream) subjected to isothermal and non-isothermal storage temperatures were collected from previously published growth curves. Maximum specific growth rate (1/h) and lag phase duration (h) were described as a function of storage temperature using the direct two-step, direct one-step and inverse dynamic modelling approaches based on various meta-heuristic optimization algorithms. The fitting capability of the modelling approaches and employed optimization algorithms was separately compared, and the one-step modelling approach for the direct methods and the Bayesian optimization method for the used algorithms provided the best goodness of fit results. These two were then further processed in validation step. The inverse dynamic modelling approach based on the Bayesian optimization algorithm yielded satisfactorily statistical indexes (1.02 > Bias factor > 1.09 and 1.07 > Accuracy factor > 1.13), which indicates it can be reliably used as an alternative way of describing the growth behaviour of Pseudomonas spp. in fish in a fast and efficient manner with minimum labour effort.

Decision Support System in Dynamic Pricing of Horticultural Products Based on the Quality Decline Due to Bacterial Growth

A decision support system (DSS) was developed to help reduce food waste at traditional food retailers while selling fresh horticultural products, but also to promote food safety and quality. This computational tool includes two major functions: (1) the prediction of the remaining shelf life of fresh horticultural product, namely lettuce, onion, carrot, and cabbage based on its microbial growth status, governed by extrinsic and intrinsic parameters (temperature, water activity and pH, respectively). The remaining shelf life of the studied horticultural products is determined by using the online predictive food microbiology tool— the Combined Database for Predictive Microbiology (Combase). The time to reach the infectious doses of bacteria considered in the study for each of the four horticultural products are predicted; (2) the calculation of the dynamic price of the produce that should be set each day, depending on the predicted end of the marketing period to increase the demand and potential for sale to the final consumer. The proposed dynamic pricing model assumes a linear relation with the remaining shelf life of the analyzed vegetable to set the selling price. The shelf life determined by the DSS for optimal storage conditions is, in general, conservative, ensuring food safety. The automatic dynamic pricing gives new opportunities to small retailers to manage their business, fostering profit and simultaneously contributing to reduce food waste. Thus, this decision support system can contribute to the sustainable value of reducing food waste by providing information to small grocers and retailers on the safety of their perishable status depending on storage conditions and allowing them to suggest a fair price depending on that quality.

Tom McMeekin: a pioneer in ‘Predictive Food Microbiology’ and Antarctic microbial ecology

Tom McMeekin: a pioneer in ‘Predictive Food Microbiology’ and Antarctic microbial ecology

A Label-Free, Non-Intrusive, and Rapid Monitoring of Bacterial Growth on Solid Medium Using Microwave Biosensor.

Microwave resonator sensors are attractive for their contactless and label-free capability of monitoring bacterial growth in liquid media. This paper outlines a new label-free microwave biosensor based on a pair of planar split ring resonators for non-invasive monitoring of bacterial growth on a solid agar media. The sensor is comprised of two split ring resonators with slightly different resonant frequencies for differential operation. The transmission coefficient (S21) of the sensor is considered as the sensor's response with a designed and measured quality factor above 200 to ensure a high-resolution operation of the biosensor. Two resonant frequencies of 1.95 and 2.11 GHz represent the sensing signal and the reference signal, respectively. The developed sensor demonstrates high performance in monitoring the growth dynamics of Escherichia coli (E. coli) on Luria-Bertani (LB) agar with 4 mm thickness. The sensor's resonant amplitude response demonstrated 0.5 dB variation corresponding to the bacterial growth over 48 hours when bacteria were spread on LB agar starting with initial OD600 = 1.5. Moreover, 0.6 dB change in the sensor's response was observed over 96 hours of bacterial growth starting with an initial OD600 = 1.17 spotted on LB agar. The measured results fit well to the curves created using Richards' bacterial growth model, showing the strength of the sensor as a potential candidate for use in predictive food microbiology systems.

Modeling Growth of Listeria and Lactic Acid Bacteria in Food Environments.

Predictive food microbiology models can facilitate the assessment and management of microbial food safety. Importantly, the combined effect of storage conditions and product characteristics can be predicted by successfully validated models. This makes it easier and faster to develop or reformulation safe food recipes and predictions can be used to documents safety of available foods. The effect of various product characteristics and storage conditions must be taken into account and extensive mathematical models including the effect of these environmental factors are needed. Here the development, evaluation and application of an extensive growth and growth boundary model for Listeria monocytogenes including the effect of 12 environmental factors as well as the growth dampening effect of lactic acid bacteria is described. The Food Spoilage and Safety Predictor software is used to illustrate how predictions can be applied.

Individual cell heterogeneity in Predictive Food Microbiology: Challenges in predicting a “noisy” world

Gene expression is a fundamentally noisy process giving rise to a significant cell to cell variability at the phenotype level. The phenotypic noise is manifested in a wide range of microbial traits. Heterogeneous behavior of individual cells is observed at the growth, survival and inactivation responses and should be taken into account in the context of Predictive Food Microbiology (PMF). Recent methodological advances can be employed for the study and modeling of single cell dynamics leading to a new generation of mechanistic models which can provide insight into the link between phenotype, gene-expression, protein and metabolic functional units at the single cell level. Such models however, need to deal with an enormous amount of interactions and processes that influence each other, forming an extremely complex system. In this review paper, we discuss the importance of noise and present the future challenges in predicting the “noisy” microbial responses in foods.

Analysis of the variability in microbial inactivation by acid treatments

The variability in microbial inactivation through acid treatments was evaluated in the present study. Enterococcus faecalis, Listeria innocua, Salmonella enterica, and Pseudomonas fluorescens were inoculated in buffered peptone water, chicken soup, and citrate solution, and were subjected to acid treatments of various intensities to reach a microbicidal effect of 0, 2, 3, and 4 logarithmic cycles. The variability in the number of survivors was greater in treated than in untreated samples. Furthermore, the effect of acid treatment on survival variability depended on the intensity of the acidification. More specifically, as the intensity of the applied acid treatment increased, the number of viable microorganisms, although smaller, was more variable. The results of this study indicate that the inactivation behaviour of microbial cells within a population is subject to variation; such variability must be quantified and taken into account in predictive food microbiology, and it can be valuable for risk assessment purposes when acidification of food is involved.

Recent trends in non-invasive in situ techniques to monitor bacterial colonies in solid (model) food.

Planktonic cells typically found in liquid systems, are routinely used for building predictive models or assessing the efficacy of food preserving technologies. However, freely suspended cells often show different susceptibility to environmental hurdles than colony cells in solid matrices. Limited oxygen, water and nutrient availability, metabolite accumulation and physical constraints due to cell immobilization in the matrix, are main factors affecting cell growth. Moreover, intra- and inter-colony interactions, as a consequence of the initial microbial load in solid systems, may affect microbial physiology. Predictive food microbiology approaches are moving toward a more realistic resemblance to food products, performing studies in structured solid systems instead of liquids. Since structured systems promote microbial cells to become immobilized and grow as colonies, it is essential to study the colony behavior, not only for food safety assurance systems, but also for understanding cell physiology and optimizing food production processes in solid matrices. Traditionally, microbial dynamics in solid systems have been assessed with a macroscopic approach by applying invasive analytical techniques; for instance, viable plate counting, which yield information about overall population. In the last years, this approach is being substituted by more mechanistically inspired ones at mesoscopic (colony) and microscopic (cell) levels. Therefore, non-invasive and in situ monitoring is mandatory for a deeper insight into bacterial colony dynamics. Several methodologies that enable high-throughput data collection have been developed, such as microscopy-based techniques coupled with image analysis and OD-based measurements in microplate readers. This research paper provides an overview of non-invasive in situ techniques to monitor bacterial colonies in solid (model) food and emphasizes their advantages and inconveniences in terms of accuracy, performance and output information.

Comparison of Primary Models to Predict Microbial Growth by the Plate Count and Absorbance Methods.

The selection of a primary model to describe microbial growth in predictive food microbiology often appears to be subjective. The objective of this research was to check the performance of different mathematical models in predicting growth parameters, both by absorbance and plate count methods. For this purpose, growth curves of three different microorganisms (Bacillus cereus, Listeria monocytogenes, and Escherichia coli) grown under the same conditions, but with different initial concentrations each, were analysed. When measuring the microbial growth of each microorganism by optical density, almost all models provided quite high goodness of fit (r 2 > 0.93) for all growth curves. The growth rate remained approximately constant for all growth curves of each microorganism, when considering one growth model, but differences were found among models. Three-phase linear model provided the lowest variation for growth rate values for all three microorganisms. Baranyi model gave a variation marginally higher, despite a much better overall fitting. When measuring the microbial growth by plate count, similar results were obtained. These results provide insight into predictive microbiology and will help food microbiologists and researchers to choose the proper primary growth predictive model.

Bacterial economics: Adaptation to stress conditions via stage-wise changes in the response mechanism

Common features of microbial adaptation are analysed with mathematical models and extended to stress conditions when the bacterial population declines before growing again. A parallel is drawn between bacterial and human communities in terms of non-mutation-based adaptation (acclimation) to stress. For a case study, the behaviour of Escherichia coli under osmotic stress, is detailed. It is suggested that stress modelling adaptation should be the focus of further developments in predictive food microbiology.

PMP 모델을 활용한 시판 Salad의 Short-term Temperature Abuse 시 미생물학적 유통기한 예측에의 적용성 검토

The aim of this study was to investigate the growth of aerobic bacteria in fresh-cut salad during short-term temperature abuse (temperature for 1, 2, and 3 h) for 72 h and to develop predictive models for the growth of total viable cells (TVC) based on Predictive food microbiology (PFM). The tool that was used, Pathogen Modeling program (PMP 7.0), predicts the growth of Aeromonas hydrophila (broth Culture, aerobic) at pH 5.6, NaCl 2.5%, and sodium nitrite 150 ppm for 72 h. Linear models through linear regression analysis; DMFit program were created based on the results obtained at 5, 10, 20, and for 72 h ( >0.9). Secondary models for the growth rate and lag time, as a function of storage temperature, were developed using the polynomial model. The initial contamination level of fresh-cut salad was 5.6 log CFU/mL of TVC during 72 h storage, and the growth rate of TVC was shown to be 0.020~1.083 CFU/mL/h ( >0.9). Also, the growth tendency of TVC was similar to that of PMP (grow rate: 0.017~0.235 CFU/mL/h; $r^2

The impact of experiment design on the parameter estimation of cardinal parameter models in predictive microbiology

In predictive food microbiology, cardinal parameter models are often applied to describe the effect of temperature, pH and/or water activity on the microbial growth rate. To identify the model parameters, full factorial designs are often used, in spite of the high experimental burden and cost related to this method.In this work, the impact of the selected experimental scheme on the estimation of the parameters of the cardinal model describing the effect of temperature, pH and/or water activity has been evaluated. In a first step, identification of a simple model describing only the effect of temperature, pH or water activity was considered. The comparison of an equidistant design and a D-optimal (based) design showed that the latter, which is based on the model’s sensitivity functions, yields more realistic parameter estimates than the typical equidistant design. By selecting the experimental levels based on the sensitivity functions, a more realistic description of the behavior around optimal conditions can be obtained.In the second step, focus was on the efficient and accurate estimation of the ten parameters of the extended cardinal model that describes the combined effect of temperature, pH and water activity on the microbial growth rate. Again, equidistant level selection is compared to a D-optimal (based) experimental design. In addition, a full factorial and a Latin-square approach are evaluated. From the simulation case studies presented, it can be stated that all parameters can be equally well defined from an equidistant design as from a D-optimal-based design. In addition, reducing the experimental load by constructing a Latin-square design does not hamper the parameter estimation procedure. This work confirms the observation of a previous study, i.e., for complex cases a Latin-square design is an attractive alternative for a full factorial design as it yields equally accurate and reliable parameter estimates while reducing the experimental workload.

A novel method for high-throughput data collection in predictive microbiology: Optical density monitoring of colony growth as a function of time

Recently, the focus of predictive food microbiology has shifted towards more mechanistically-inspired modelling. Together with this trend, the need for methods that allow rapid data collection at the (intra)cellular level, as well as the intermediate subpopulation/colony level, has emerged. Although several experimental techniques are currently available to study colony dynamics in/on solid media, their widespread implementation as high-throughput methods remains a challenge.In this research, a novel method is presented to study colony growth based on optical density measurements performed in microtiter plates. An area scan procedure was applied to monitor individual Escherichia coli colonies in 48-well plates at 30 °C. Based on a fixed threshold value to separate the object (colony) from the background, the colony area was determined as a function of time.With this technique, expansion of the colony in radial direction could be monitored. Practical limitations (i.e., maximum achievable resolution and colony size) of the proposed method were investigated. A comparison was made with existing methods at the level of hardware requirements, data acquisition and data processing. Overall, the novel optical density method proved to be a flexible, high-throughput tool for monitoring (the mechanisms of) microbial colony growth in solid(like) systems.

Comparing experimental design schemes in predictive food microbiology: Optimal parameter estimation of secondary models

In predictive food microbiology, full factorial designs are still more the rule than the exception, despite the huge experimental workload and cost related to this method. In this study, two simulation studies for secondary square-root-type models are performed to compare several experimental designs with respect to four criteria: (i) number of experiments, (ii) goodness-of-fit statistics with respect to the original model structure, and (iii) accuracy and (iv) uncertainty of the parameter estimates. In addition, the effect of data quality, quantified as the error related to plate count measurements, is assessed on the relation between model structure and experimental design. Full factorial, reduced full factorial, central composite, Latin-square and Box-Behnken designs are evaluated and compared to randomly selected datasets.As a guideline, a full factorial design should be preferred for rather simple model structures and a limited number of levels per environmental factor. For more complex cases, a Latin-square design is an attractive alternative as it does not require a priori model knowledge and provides relatively accurate and reliable parameter estimates while keeping the experimental efforts to a minimum.

Predictive modelling of Salmonella: From cell cycle measurements to e-models

The quantitative measurements of the growth of Salmonella can be traced back to the 1950s, when the Copenhagen School studied its cell cycle. Although predictive food microbiology has only been recognised as a discipline in its own right since the 1980s, the first predictive models on Salmonella, specifically on its thermal inactivation, were published in the 1960s. Today this is the foodborne pathogen for which the most D-values can be found in the literature. Being of concern in meat, growth models were developed mainly in poultry, other meats, egg products and culture media. With the advent of the internet, predictive modelling has become more advanced in terms of organising, analysing and visualising large amounts of data, and it has also become easier to disseminate the resultant predictive software packages. We anticipate that computational developments will generate further improvements, including complex scenario analysis, probabilistic and dynamic models.

APPLICATION OF ARTIFICIAL NEURAL NETWORKS TO PREDICT CLOSTRIDIUM BOTULINUM GROWTH AS A FUNCTION OF ZATARIA MULTIFLORA ESSENTIAL OIL, pH, NaCl AND TEMPERATURE

ABSTRACT The present study is undertaken to predict the effects of Zataria multflora essential oil, NaCl, temperature and pH on the probability percentage of growth of Clostridium botulinum using artificial neural network (ANN). To assess the effects of essential oil, NaCl, pH and temperature on Log P% of C. botulinum type A spores, the experiment was arranged in a factorial design. This design (4 × 3 × 3 × 3) included four levels of EO (0.0, 0.015, 0.03 and 0.06%), three levels of NaCl (0.5, 4 and 6%), three levels of pH (6, 6.5 and 7.4), three storage temperatures (15, 25 and 35C) and repeated observations (12 times) for growth in BHI broth for up to 43 days (D). According to our results, the log P% of C. botulinum was affected significantly (P < 0.01) by Essential oil, pH, NaCl and temperature. The numbers of neurons in input hidden and output layers were 5, 11 and 1, respectively. The inputs to the ANN were temperature, pH, NaCl, essential oil and the number of days after inoculation. The output of the ANN was log P%. Our results show that the ANN is a suitable model for predicting the growth of C. botulinum when it is affected by different levels of pH, temperature, NaCl and essential oil.PRACTICAL APPLICATIONS Clostridium botulinum is an anaerobic bacterium that produces a powerful neurotoxin. Because there is concern over the safety of some chemical preservatives, an increased interest is started into more natural alternatives. Zataria multiflora has been used traditionally as flavor agent and its antimicrobial activities has been established in previous studies. In addition, predictive food microbiology involves the knowledge of microbial growth responses to environmental factors summarized as mathematical models. Over the last few years, artificial neural networks (ANNs), as nonlinear modeling techniques, have been proposed for use in predictive microbiology. Considering the importance of C. botulinum and using natural preservative in food safety and the power of ANN in predicting purposes, we suggest that ANN can be used as a prediction tool in food safety.