Clin Nutr. 2025 Oct 28;55:11-23. doi: 10.1016/j.clnu.2025.10.009. Online ahead of print.
ABSTRACT
Recent advances in artificial intelligence, wearable biosensors, and multi-omics technologies, combined with interdisciplinary translational collaborations, are transforming the landscape of nutrition, particularly by enabling precision nutrition. We propose computational nutrition as an emerging interdisciplinary field that aims to address complex challenges in nutrition and health and drive paradigm shifts in nutrition research through computational methods in statistics and computer science (e.g., statistical modeling, simulation, causal inference, machine learning, and deep learning) and multi-modal data. Computational nutrition employs epidemiology and data science as its methodologies and integrates expertise from nutrition, food science, computer science, statistics, systems biology, and public health. The main research directions of computational nutrition are: 1) prediction of personalized metabolic responses to foods and establishment of individualized dietary reference intakes; 2) causal inference in nutrition and diseases and evaluation of individualized treatment effects of nutritional interventions; 3) precise and dynamic assessment and monitoring of diet-related disease risks; 4) simulation and evaluation of public health nutrition policies and sustainability assessment of dietary patterns. Critical challenges such as the reliability of wearable biosensors, trade-offs in feature selection, ethics of algorithms and health equity, and interpretability of algorithms are raised, which must be addressed to ensure human well-being and rights.
PMID:41176812 | DOI:10.1016/j.clnu.2025.10.009
