Early detection of food risks with AI and Big Data
Emerging risks in the food chain are increasingly impacting our daily lives. How can food industry and authorities leverage technologies like AI and Big data in the future to improve both the safety, healthiness and sustainability of our food? The new EU project HOLiFOOD answers this question by using AI and Big Data to develop early warning systems that will protect consumers from food safety risks. It will transform, the risk analysis framework from reactive to proactive, to support the transition to a sustainable and secure food system.
"The HOLiFOOD project aims to enhance the framework for food safety risk analysis, harnessing the power of artificial intelligence and big data. By doing so, it will uphold Europe's high standards of food safety, while promoting secure and sustainable food production," said Prof. Dr. Ine van der Fels, coordinator of HOLiFOOD from Wageningen University & Research.
Food systems urgently need a holistic approach
As food systems are threatened globally by unprecedented challenges, there is an urgent need to transform the food system to deliver sufficient, affordable and healthy food for all. New tools such as the risk–benefit assessment (RBA) have been developed in recent years to provide guidance on dietary habits and establish new policies (EFSA, 2010) while ensuring food safety and protecting consumers against emerging hazards.
The HOLiFOOD project will integrate risk-benefit assessment and cost-benefit considerations into one 'holistic' framework for risk analysis. Combined with AI and big data tools this framework will support early identification of emerging food safety hazards in the food system. This will allow food safety authorities to make better decisions by considering the whole environment (societal, environmental, economic) within which food is being produced.
Cereals, legumes and poultry– how can we guarantee food safety across these food chains?
Three selected supply chains—cereals (maize), legumes (lentils) and poultry (broiler meat)—will be used to study emerging risks using AI and Big Data technologies. These three food chains have been chosen using two criteria: maize and poultry, for their importance related to production and consumption throughout Europe, whereas the lentils supply chain was selected as a small food crop that is expected to grow due to its potential as a plant-based protein alternative.
Using state-of-the-art technologies, HOliFOOD partners will develop new and improved methods for detecting known and unknown chemical and biological hazards (e.g., bacteria, viruses) across the three selected supply chains. The project will also strive to create generic methods and tools applicable to the entire food system.
Co-designing new methods to meet real needs of end-users
Adoption of the methods and results would not be possible without understanding individual needs of end-users of these technologies, ensuring that information about emerging food safety risks is effectively exchanged with the public, and that what is being done to prevent and mitigate the risks is understood.
That is why HOLiFOOD will take a holistic perspective on its stakeholder community, wherein specific activities will be targeted to understanding individual information needs and addressing adoption barriers. Project partners, among them researchers, NGOs, policymakers and companies, will engage stakeholders and citizens during all phases of the project—development, testing and implementation of the models and methods.
End-users will help design and test the project's methods in spaces known as 'living labs', which the European Commission (EC) describes as 'user-centered, open innovation ecosystems'—a new approach that warrants integration of research and innovation processes in real life communities and settings. Here, the citizen is placed at the center of innovation, allowing individual stakeholders to be considered throughout the lifecycle of the project.
More information:
HOLiFOOD project: holifoodproject.eu/
Provided by European Food Information Council