Fighting overweight risks with digital lifestyle recommenders
A tailored approach aims to prevent 10,000 cases of obesity per year
Over 30% of Europe’s citizens in vulnerable populations are at risk of becoming overweight or even obese. Many current interventions to address this problem fail because they don't consider personal factors like socioeconomic status, biology, and environment. Additionally, they often overlook emotional aspects and fail to engage and motivate the user.
Tailored to each user
The HealthyW8 research project takes a different approach. It will address these shortcomings by creating a digital-based lifestyle recommender. The recommender is tailored to each user and incorporates a human digital twin to bridge the gap between science, society, and stakeholders. The project will focus on schoolchildren, young adults, and the elderly. It estimates to be able to prevent 10,000 cases of obesity per year with 200,000 users. The long-term goal is to maximize impact by having EU institutions adopt HealthyW8's methodology and tools.
The project is led by by the Luxembourg Institute for Health. Our IE&IS colleague Pieter Van Gorp is one of the TU/e experts in HealthyW8. He will focus on the design of the information system infrastructure for the lifestyle recommender system. In particular, he will supervise a PhD student who will study how engaging content can be generated responsibly through the controlled use of large language model technology. Especially considering the vulnerability of the target group, special measures will have to be taken to ensure that the AI-generated content satisfies quality and safety constraints. Such research contributes to more responsible forms of AI. Van Gorp will work closely with Astrid Kemperman, from the department of the Built Environment. Kemperman will also supervise a PhD student, who will focus more on the experimental design and theoretical grounding of the behavior change techniques.
This research is being conducted as part of the HealthyW8 project which received funding from the European Union's Horizon Europe Research and Innovation Programme under the grant agreement n°101080645.