The United Nations states “End hunger, achieve food security and improved nutrition and promote sustainable agriculture" as one of its sustainable development goals by the target date of 2030. To achieve these goals, global food and agriculture systems will require profound changes, in which big data and AI technologies can play significant roles.

In the past decade, a huge amount of work has been done in biomedical predictive modelling. While there are extensive resources available for the biomedical domain, the food and nutrition domains are relatively low AI-resourced. In the last five years, several studies have been focused on relevant food information extraction from textual data, food image recognition, food portion estimation, food ontology alignment, etc. Despite the recent studies focused on food and nutrition data, there is a requirement for additional effort in order to provide resources that can extract useful information from heterogeneous food and nutrition data sources. Developing such AI resources will allow further linking food and nutrition data with biomedical and environmental data, which can help explore heterogeneous hidden relations.

The focus on this track is to provide an overview of AI methods that have already existed for food and nutrition data, together with methods for linking biomedical research data with food and nutrition data as well as on methods that address key challenges arising in application areas relevant to personalized nutrition and medicine, and food safety and traceability.

This full-day track, will consist of two parts. The first part will focus more on AI methods and resources used for analyzing food and nutrition data with regard to ​personalized nutrition and medicine​, while the second part will be focused on AI analysis for ​food safety and traceability​.