The large Horizon Europe project will trial innovative diet-monitoring technologies and develop AI tools to deliver personalized nutritional advice.
Heart disease, diabetes or obesity. These are some of the common non-communicable diseases (NCD) that our society has been facing for a long time. The international CoDiet project supported by the prestigious Horizon Europe program will test a new approach to their prevention.
Experts from 17 institutions in 10 countries will use innovative tools to monitor and evaluate the eating habits of patients in order to better understand the relationship between food and diseases. The output of the project will be an AI-based system, which will offer people an effective, tailored nutritional plan. Scientists from our AI Center will contribute to this personalized nutritional advice by developing optimization algorithms.
Using AI to understand the link between diet and disease
An unhealthy diet is associated with metabolic changes and an increased risk of serious non-communicable diseases. According to the World Health Organization (WHO), these diseases kill 41 million people every year, which is equivalent to 74% of all deaths worldwide. However, we still know very little about the specific dietary mechanisms that actually cause disease.
Additionally, current tools used to collect dietary information rely on user input, which can be unreliable. Research to date also lacks data on vulnerable groups, such as people from lower socio-economic backgrounds, among whom NCDs are often over-represented (source).
Personalized nutrition plans for everyone
The CoDiet project led by the Spanish research center AZTI will address these deficiencies. Its goal is to develop a tool based on artificial intelligence that can evaluate the individual risk of diet-related diseases and provide nutritional advice tailored to the individual user.
"It is well known that each person's metabolic response to the same diet varies. CoDiet will work on personalizing dietary advice instead of a "one-size-fits-all" approach," said research leader dr. Itziar Tueros from AZTI and added that for such a solution it was necessary to assemble a multidisciplinary team.
Intelligent camera monitoring our food intake
One of the main gaps in our knowledge is understanding exactly what people consume in their daily lives. Current tools are very imprecise, which complicates the understanding of the relationship between diet and disease. The project will therefore test an intelligent wearable camera developed at Imperial College London.
It is designed to be comfortably worn on the ear and passively record what the wearer consumes. It uses computer vision and deep learning to collect and analyze data, thanks to which it automatically recognizes food types and approximate portion sizes. This system will be supplemented with other technologies that will help to understand the processing of food in the body, including the analysis of the intestinal microbiome and metabolites in urine.
Optimization algorithms developed at CTU
Jakub Mareček together with teams at the Technion Israel Institute of Technology, the National and Kapodistrian University of Athens and Imperial College London are working on methods for learning causal relationships from data.
Hormones and food-cravings
"Does a particular hormone affect what we crave, or does our diet affect the concentration of a particular hormone? This is the question we are hoping to adequately answer at the level of algorithms. Machine Learning revealing causality is a big open problem in both statistics and artificial intelligence, and our new optimization-based methods should contribute to solving it," explained Marecek.
Our researchers are pleased that the modern optimization methods they are developing can help solve these fundamental social problems.