“Growth Predictor”: A new predictive modelling and quantitative microbial risk assessment tool
Year: 2025
Type: article
Abstract: A new predictive modelling and quantitative microbial risk assessment (QMRA) software, developed in R, is available on-line ( https://skandamis.shinyapps.io/Microbial-Growth-Predictor-Dashboard/ ). Primary model fitting is carried out with the Baranyi model. Secondary fitting and growth simulations are based on gamma models with or without interactions. The same gamma terms can be used for fitting and growth simulations under static or dynamic conditions. One of the novel features is the use of normal distributions to describe the variability in T, pH, a w, the levels of a single inhibitor and the inter-strain variability in growth limits. The QMRA is comprised of four consecutive modules from primary production until consumption. In addition to prevalence, the modules may also consider partition, mixing and cross-contamination. Variability can be introduced through a variety of probability distributions, for initial contamination, re-contamination, storage time and temperature, product characteristics, serving size and maximum population density. Fixed or variable log reductions during cooking, may be introduced as user-defined values or probability distributions, respectively, or estimated by a Bigelow thermal inactivation model. The trilinear primary growth model is used for estimating log changes, based on μ max obtained by gamma models. The QMRA outputs include graphical distribution of ingested dose and probability of illness ( P ill ), as well as tabular estimates. The user may select built-in dose-response models, or define the parameters values of exponential, beta-Poisson, beta-binomial and binomial dose-response models. The tool is intended for multiple users from the scientific community, the authorities and the food industry, for growth simulations and high-resolution risk assessments. • Growth Predictor is a fitting, growth simulation and modular risk assessment software. • Gamma models for estimating cardinal values and simulating static or dynamic growth. • Variability in growth conditions and cardinal values is embedded in simulations. • Uncertainty in lag time is considered in growth simulation.
Source: Food Research International
Author: Panagiotis N. Skandamis
Institution: Agricultural University of Athens
Language: English
Topic: Food Safety and Hygiene
Subfield: Food Science
Domain: Life Sciences
Open Access status: hybrid
Estimated APC paid: $4,320
Award: G8814088819
OpenAlex ID: W4409202033