The harmonization of terms and concepts as well as the generation of information exchange formats are urgent issues in different domains of predictive modelling in food.
Before the actual technical work can be started it would be helpful to agree within the community on a general strategy. This is a first proposal for such a strategy including suggestions on decisions that should be taken early on:
Thesis 1:
The practical work should be organized as bottom-up approach, meaning that harmonization efforts in different sub-domains can evolve independently as long as they comply to a common goal of higher-level integration. As first sub-domains I propose: "Units and terminologies", "predictive microbial modelling", "source attribution modelling", "tracability and network modelling" and "lab data description".
The domain of "QMRA modelling" could then serve as common goal of higher-level integration as is nicely visualized in the ICRA online tool. In a long-term perspective it is even possible to stretch out to the domain of "chemical risk assessment modelling".
This approach could be technically supported by establishment of sub-domain specific forums within the Discussion board.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I support the bottom-up approach, since it allows to concentrate on one question at a time.
Due to the interdisciplinary nature of this project, the definition of the above mentioned sub-domains may not be clear for those who are coming from an IT background, like myself.
Can you please provide a short description about the given sub-domains, what they are about, what are the challenges?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Units and terminologies
This sub-domain could discuss on how to harmonize the description of units of measurements and all sorts of entities that are considered in a risk assessment or food safety model, e.g. organisms, food matrices, environmental factors etc..
predictive microbial modelling
This sub-domain specifically addresses the issue of an information exchange format for "classical" predictive microbiology.
source attribution modelling
This sub-domain could address issues relevant for source attribution models. As far as I know this is strongly related to surveillance and serotyping data, classical statistical as well as Bayesian modelling approaches.
traceability and network modelling
This sub-domain addresses issues relevant to food distribution data and models, e.g. spatio-temporal data on the distribution of food items provided by wholesalers and / or food producers, gravity models, transportation models etc. These data and models could e.g. be used to help tracing back food contaminations and /or to create reasonable risk estimates in outbreak situations.
lab data description
This sub-domain could be used to specifically address issues relevant for a scientific sound documentation of raw data measurements.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The harmonization of terms and concepts as well as the generation of information exchange formats are urgent issues in different domains of predictive modelling in food.
Before the actual technical work can be started it would be helpful to agree within the community on a general strategy. This is a first proposal for such a strategy including suggestions on decisions that should be taken early on:
Thesis 1:
The practical work should be organized as bottom-up approach, meaning that harmonization efforts in different sub-domains can evolve independently as long as they comply to a common goal of higher-level integration. As first sub-domains I propose: "Units and terminologies", "predictive microbial modelling", "source attribution modelling", "tracability and network modelling" and "lab data description".
The domain of "QMRA modelling" could then serve as common goal of higher-level integration as is nicely visualized in the ICRA online tool. In a long-term perspective it is even possible to stretch out to the domain of "chemical risk assessment modelling".
This approach could be technically supported by establishment of sub-domain specific forums within the Discussion board.
I support the bottom-up approach, since it allows to concentrate on one question at a time.
Due to the interdisciplinary nature of this project, the definition of the above mentioned sub-domains may not be clear for those who are coming from an IT background, like myself.
Can you please provide a short description about the given sub-domains, what they are about, what are the challenges?
Units and terminologies
This sub-domain could discuss on how to harmonize the description of units of measurements and all sorts of entities that are considered in a risk assessment or food safety model, e.g. organisms, food matrices, environmental factors etc..
predictive microbial modelling
This sub-domain specifically addresses the issue of an information exchange format for "classical" predictive microbiology.
source attribution modelling
This sub-domain could address issues relevant for source attribution models. As far as I know this is strongly related to surveillance and serotyping data, classical statistical as well as Bayesian modelling approaches.
traceability and network modelling
This sub-domain addresses issues relevant to food distribution data and models, e.g. spatio-temporal data on the distribution of food items provided by wholesalers and / or food producers, gravity models, transportation models etc. These data and models could e.g. be used to help tracing back food contaminations and /or to create reasonable risk estimates in outbreak situations.
lab data description
This sub-domain could be used to specifically address issues relevant for a scientific sound documentation of raw data measurements.