The Basque technology center AZTI has launched research to study the relationship between diets and the most common diseases using the latest technologies, such as portable smart cameras
Unhealthy diets are associated with metabolic changes and an increased risk of diseases such as heart disease, diabetes, obesity either cancer. In the Basque Country, the AZTI technology center has launched a research project to study the relationship between diets and this type of ailments by testing innovative technologies, as wearable smart camerasor the development of tools of artificial intelligence to offer personalized advice, as explained to El Periódico de España, from the Prensa Ibérica group, Itziar Tueros, principal investigator.
Tueros is responsible for the department of Food and Health of the center, specialized in the marine environment and food. Noncommunicable diseases, with data from the World Health Organization (WHO), kill 41 million people each year, which is equivalent to 74% of all deaths. The researchers wanted to know more about the mechanisms connecting nutrition and these ailments.
The tools that are used today to collect information about our eating habits are based on questionnaires, which can be unreliable, review. There is also a lack of data on vulnerable groups, such as those of lower socioeconomic status, in which these diseases are usually poorly represented, indicate the authors of the work. The CoDiet project, coordinated by AZTI and funded by the European Union and UK Research and Innovation (UKRI), aims to fill these gaps.
reliable answers
“Currently we use questionnaires where the person is asked about what they have eaten in the last days or the general dietary habit. For example: ‘How many times a week do you eat fish, or eggs?’. But they are imprecise because many times it depends on the memory of the person, or it is difficult to know how much of each food we eat Or we are simply not interested in verbalizing that we eat a lot of one type of food and little of another. We need to have methods that give us more objective information like the ones we are going to test in this project”, explains the researcher Iztiar Tueros.
In addition, Tueros indicates that there is “a greater frequency of non-communicable diseases in the population with a low socioeconomic level, due to less healthy lifestyles, either due to scarce resources or due to lower educational level“. There are studies, he adds, that point out that in this strip there is a greater consumption of alcohol and tobacco and lower consumption of fruits, vegetables, fish and fiber than in populations with high socioeconomic level.
smart camera
Your team’s work will develop a bespoke AI-based tool capable of assessing individual risk of diet-induced non-communicable diseases. The project builds on previous research by the imperial college londonwhich includes the development of a wearable smart camera and personalized nutrition.
“Imperial College London developed a prototype of an intelligent camera that is placed on the ear and that is capable of capturing what we eat during the period that we want to study,” says Tueros.
“Imperial College London developed a smart camera prototype that is placed in the ear and that is capable of capturing what we eat during the period that we want to study. Then, through artificial intelligence, the plate is recognized, distinguish if they are lentils or beans and in what quantity and, finally, this information is transformed into nutrients. Previous results indicate that it is quite accurate when it comes to capturing nutritional habit, that’s why now we want to test it on a larger scale“, details the researcher.
dietary information
The camera is one of the tools collection of dietary information that the project will test. will use novel computer vision techniques and deep learning to automatically recognize food types and estimate portion sizes. This will be combined with other technologies that help them understand how food is processed in the body, including gut microbiome analysis and the metabolites in the urine.
The analysis of metabolites in urine allows us to know if we eat a type of vegetable meat protein or if we have a varied diet in terms of types of fruits and vegetables, fibers… etc
He analysis of metabolites in urine, says Tueros, allows us to know if we eat a type of vegetable meat protein or if we have a varied diet in terms of to types of fruits and vegetables, fibers…etc. He of the microbiome in feces, know if we have a composition of microorganisms that makes it easier for us the metabolism of certain nutrients such as some vitamins; the analysis of lipids in red blood cells informs us about the type of fats we eat and their metabolic functions. “These solutions, complement the smart camera and they allow us to objectify our food intake and we will test them in the CoDiet project in four countries”, he adds.
The one size fits all
The project will also develop a tool that will simulate the change in diseases in response to diet at the population level. The objective is to promote that diets are followed that protect against these ailments. Because, the researchers warn, right now tackling these problems is based on a “one size fits all” approach. They trust that personalized nutritional advice can give more effective results. The Basque researcher explains that each person has a different metabolic response to the same diet.
The technological tools will be tested on a total of 700 people spread over four countries
“All these technological tools that we will test in a total of 700 people distributed in four countries, will allow us to generate a lot of data of different kindsphysical activity, body composition, metabolism, behavior”, he explains. They will also use a tool that allows generating some simulation models of health policies, where we will be able to compare different scenarios of nutritional recommendations and evaluate the impact they would have at the health or economic, adds.
18 universities and research centers from Ireland, Germany, Belgium, Greece, Estonia, Italy, Israel, the United Kingdom, the Czech Republic and Spain are collaborating in this work – which began in 2023 and will last four years (AZTI, CIC bioGUNE , University of Valencia, Microcaya and the Consortium Center for Biomedical Research in Red-CIBER). The researchers hope to have the final results in 2026-2027, concludes Tueros.