1. HOLiFOOD responds to the need to develop new methods for Early Warning (EW) and Emerging Risk Prediction (ERI) and applies artificial intelligence (AI) and ‘big data’ technologies to automatically extract data relevant to drivers of food safety risks from publicly available websites and databases, and integrates these in AI ERI prediction models. Thus far, various AI-based tools have been developed for text mining and analysis, and prediction. Large amounts of data relevant to food safety have been gathered, cleaned, and pooled into a “large CompreHensive European Food Safety database” (CHEFS).
2. HOLiFOOD develops and validates methods and devices for the identification and characterization of existing and (re-)emerging hazards with the aim of anticipating and possibly mitigating/preventing impacts. It develops an untargeted pathogen detection procedure that combines non-selective enrichments, metagenomics analyses, and proof of concept for the poultry and legume chains. Several other devices for on-site analysis are being developd. Deep Learning algorithms based on AI are developed for feature detection in MS Data from Liquid Chromatography-High Resolution Mass Spectrometry (LC-hrMS).
3. Holistic risk assessment for regulation. Focus is on embedding food safety risks in a comprehensive cost-benefit analysis of the food system including positive and negative health, environment and economic dimensions. Different scenarios are considered for lentils, maize, and poultry in several EU countries, with climate change as the main driver of change. A general framework for holistic risk assessment is being developed, and is planned to be finished by the end of the project.
4. Data and knowledge sharing infrastructures: HOLiFOOD takes the next step in aligning and harmonizing existing work that is carried out by both public and private stakeholders, boosting it with the development of a joint data and model infrastructure of which all European stakeholders can take advantage. Focus is on creating a shared registry of datasets that may be used for food risk mitigation, by connecting and extending them. Infrastructure was designed and set up that is integrated with software systems that support food risk mitigation decisions for multiple actors in the supply chain.
5. Codesign and citizen science: to understand the food safety system, interested stakeholders, and their interests, role and influence within the system and in relation to ERI is conducted. This ensures that representation from relevant sectors and institutions are included. Citizen science is applied and Living Labs are used to ensure that all stakeholders benefit from improved food safety, including consumers in vulnerable groups, small businesses and cooperatives within local supply chains.