L3S Best Publication Award Q3+Q4/2025
Category: AI for Food Science
AI-Driven Personalized Nutrition: Integrating Omics, Ethics, and Digital Health
Authors: C. Mundt, B. Yusufoglu, D. Kudenko, K. Mertoglu, T. Esatbeyoglu
Published in Molecular Nutrition & Food Research
The paper in a nutshell:
Most people receive general dietary advice, even though individuals can respond very differently to the same foods. Our paper reviews how artificial intelligence (AI) can help address this by analyzing complex health data to better understand individual nutritional needs.
Which problem do you solve with your research?
Many chronic diseases, such as obesity, type 2 diabetes, and cardiovascular diseases, are strongly influenced by diet. However, most dietary recommendations are based on generalized guidelines and do not consider the large biological differences between individuals. As people respond differently to the same foods because of variations in genetics, metabolism, gut microbiota, lifestyle, and environmental factors, it is difficult to create dietary strategies that are effective for everyone.
What is the potential impact of your findings?
Our review shows that AI can detect patterns in health data, from genes and metabolism to the gut microbiome and lifestyle habits. By connecting all these dots, AI can create personalized dietary recommendations that adapt to each individual’s needs. This approach can revolutionize how chronic diseases are managed and prevented. In the future, AI-driven nutrition is expected to make healthy eating easier, more precise, and tailored to individual people and not just general guidelines.
What is new about your research?
Our paper not only summarizes latest developments in AI-assisted personalized nutrition but also critically examines emerging approaches and implementation challenges. On the one hand, we highlight the growing role of microbiome-based models, digital health tools, and predictive systems such as digital twins and health knowledge graphs. On the other hand, we also address key challenges, such as algorithmic bias, data privacy, and limited representation of diverse populations, and emphasize the need fair and clinically reliable AI-applications.
Link to the paper: https://onlinelibrary.wiley.com/doi/10.1002/mnfr.70293
