Activity: Talks and presentations › Conference presentations
In quantitative microbiological risk assessment (QMRA), risk estimates depend on prevalence of contamination and the microbial concentrations in food products. The probability distribution describing variability in concentrations and the method used to fit it to microbial data may therefore influence the accuracy of risk estimation. A challenge underlying the fitting is that a low concentration goes often undetected in plate counting, producing an “artificial zero” that should be differentiated from a “true” zero (from an uncontaminated product), while fitting a distribution. An accurate separation between the two types of zeroes may be essential for the characterization of prevalence and distribution of concentrations, and therefore for an accurate risk estimation.
31 Oct 2013
Advances in Predictive Modeling and Quantitative Microbiological Risk Assessment of Foods: null