Periodic Reporting for period 3 - PROTEIN (PeRsOnalized nutriTion for hEalthy livINg)
Période du rapport: 2021-06-01 au 2022-11-30
"Unhealthy diet and lack of physical activity, in conjunction with tobacco use or harmful use of alcohol, may lead to raised blood pressure, overweight/obesity, hyperglycaemia (high blood glucose levels) and hyperlipidaemia (high levels of fat in the blood), which are the main metabolic factors that significantly increase the risk of NCDs" (Source: WHO - Non-communicable diseases, 2017). Despite these risks, today's diet is characterized by irregular and poorly balanced meals. Unhealthy eating habits in our daily life are not only risk factors for non-communicable diseases, but also major causes of stress and tiredness, i.e. lack of energy. Knowledge about our dietary habits based on the analysis of different types of information, including individual parameters, e.g. physical and psychological characteristics, lifestyle, preferences etc., can contribute greatly towards answering key questions to respond to societal challenges regarding food and health. New advances in ICT technologies and, especially, in wearable sensors, big data analysis and machine learning techniques, as well as in direct-to-consumer genetic testing, blood and gut microbiome analysis, open new opportunities for researchers to monitor and collect information related to our dietary behaviours. This analysis can constitute the basis for the development of innovative solutions for personalized nutrition advice and support helping consumers achieve long term healthy and sustainable diets.
Motivated by these facts, the PROTEIN project aimed to develop an end-to-end ecosystem for engaging people to a healthy, pleasurable, nutritional, and sustainable diet by offering dietary and physical activity programs adapted to their needs and driven by their personal preferences, physical and physiological characteristics as well as their health status. Specifically, the main objective of PROTEIN is to create an ICT-based system for providing personalized nutrition based on the collection and analysis of large volumes of data related to users' dietary behavioural patterns (e.g. food choices, calories, macronutrients and micronutrients intake etc.), physical activity (motion, exercise etc) and individual parameters (electronic health record, genetic information, blood parameters, gut microbiome analysis, preferences and socio-cultural aspects). PROTEIN proposes a radically novel approach to advice and support consumers in everyday living, while ensuring users’ privacy protection i.e. data will be anonymized and securely stored in the Cloud for processing.
By creating such a novel and complex ecosystem and using it in extensive piloting activities with real users (always supervised by appropriate experts), the PROTEIN project revealed significant insight related to our dietary behaviours, our food choice drivers, and their relation to our personal profile.
The PROTEIN ecosystem consists of a mobile application that has been published in Google Play (search for 'PROTEIN EU') and a web-based Dashboard for nutritionists and other experts, and the users. The ecosystem takes advantage of recent advances in a multitude of information and communication technologies, e.g. in mobile devices, wearable biosensors, big data analysis, artificial intelligence, machine learning, direct-to-consumer genetic testing, and blood and gut microbiome analysis. It is based on novel scientific achievements and includes numerous novel technologies: ontology containing nutrition domain knowledge, image-based food identification and food weight estimation, food intake estimation based on smart scale data, eating rate analysis based on video or on smartwatch accelerometer data, bowel sound analysis using a smart belt, a novel Volatile Organic Compounds (VOC) sensor, and, ultimately, an AI-based nutrition and physical activity advisor. The PROTEIN ecosystem has been designed, developed, integrated, verified, and evaluated by project experts and real users exclusively during the EU funded PROTEIN project.
A series of concrete innovation assets have been produced by the project and are accompanied by detailed exploitation analyses: a 'VoC sensor' for non-invasive breath analysis of food intake effects, a 'Smart Belt' wearable solution for intestinal functioning assessment, an algorithm for sensor-based eating rate analysis, an algorithm for video-based eating rate estimation, an algorithm for image-based food identification and intake assessment, and a personalised nutrition/physical activity mobile application.