On April 29th 2025, the European Cluster of Obesity Research Projects (OBEClust) held its first webinar titled “AI and Obesity Prevention: Leveraging Technology to Address Obesity”. Moderated by OBEClust Chair and PAS GRAS Project Coordinator Mr. Paulo Oliveira, the event marked a strong example of collaborative action among nine EU-funded projects aiming to reshape obesity prevention through technology and data-driven innovation.
AI in Obesity: Opportunities and Challenges
Ms. Jennifer Baker, President-Elect of the European Association for the Study of Obesity (EASO) and Head of Lifecourse Epidemiology Research at Copenhagen University Hospital-Bispebjerg and Frederiksberg, opened the webinar with a broad overview of how AI is currently applied in obesity prevention and care. She emphasised the importance of both structured data (e.g., clinical trials) and organic data (e.g., social media) as “fuel” for AI systems. She highlighted predictive analytics, precision medicine, and the use of electronic health records (EHRs) as promising areas—while noting ethical, practical, and bias-related challenges. She stressed the need for multidisciplinary collaboration and the inclusion of patient perspectives.
Childhood Obesity: Predicting Risk with AI
Mr. Nicolas Gambardella from the Centre national de la recherche scientifique (CNRS) and representing the Obelisk project discussed efforts to use deep AI models to predict childhood obesity risk at birth, incorporating both biological and socio-economic data among others. He underlined the growing rates of childhood obesity across all EU countries and its long-term health and societal consequences.
Understanding Metabolic Diseases through AI
Ms. Silvia Sabatini, Junior Researcher at the Institute of Clinical Physiology of the National Research Council in Pisa presented how AI can identify heterogeneity in metabolic diseases. Using unsupervised learning and clinical data, she highlighted research aiming to uncover subtypes of disease, associated risks, and comorbidities, offering more nuanced approaches to diagnosis and treatment.
Machine Learning in Obesity Interventions
Mr. Andreas Vezakis from the National Technical University of Athens (NTUA) and representative of the BIO-STREAMS project stressed the importance of early detection and long-term strategies in childhood obesity prevention. His talk focused on how machine learning can integrate various data types, such as socioeconomic and behavioral, to improve outcomes, while also addressing key challenges like data fragmentation, ethics, and standardization.
Uncovering Obesity’s Early Determinants
Representing the eprObes project, Mr. Peter Atanasov (AI Solutions, Amaris) and Dr. Alex Bravo Serrano (Senior Researcher at Amaris) discussed how machine learning and epigenetics can uncover early biological markers for obesity. Dr. Serrano also explored literature insights and generative AI’s potential in advancing the field.
Personalised Insights with Causal AI
Associate Professor Stavroula Georgia Mougiakakou, from the Artificial Intelligence in Health and Nutrition Laboratory of the University of Bern, and member of the BETTER4U project concluded the webinar with a presentation on causal AI and its use in personalising obesity interventions. She highlighted how multimodal data can offer deeper, actionable insights into prevention strategies.
Don’t miss out!
The webinar ended with an engaging Q&A session, fostering dialogue across projects and disciplines. For more insights about OBEClust and its mission, you can access our first joint press release and project overviews here.