Artificial Intelligence in Food-Nutrition-Health Research: From Multimodal Data Integration to Precision Intervention

Area of Science:

• Food Science
• Nutrition Science
• Health Informatics
• Artificial Intelligence

Background:

• Artificial intelligence (AI) is increasingly vital for analyzing complex, high-dimensional datasets in food-nutrition-health research.
• Traditional hypothesis-driven methods struggle with the intricacies of modern food and health data.
• A systematic review is needed to synthesize AI's progress across the food-nutrition-health continuum.

Purpose of the Study:

• To systematically review and synthesize research progress on AI in the food-nutrition-health continuum from 2020 to 2025.
• To provide a comprehensive overview of AI's technical foundations, applications, challenges, and future prospects in this domain.
• To propose a framework for integrating diverse data sources and AI technologies.

Main Methods:

• Systematic review of 181 studies published between 2020 and 2025.
• PRISMA-guided selection process for identifying relevant systematic reviews.
• Analysis of AI's technical layers (data, technology, algorithms) and application scenarios.

Main Results:

• AI applications span food analysis, safety, nutrition-disease modeling, pathogen detection, and personalized nutrition.
• Key technological advancements include nondestructive testing (spectroscopy, NMR, imaging) and deep learning algorithms.
• A tripartite framework is proposed, integrating multisource data, advanced technologies, and sophisticated algorithms.

Conclusions:

• Critical challenges include model generalization, algorithmic transparency, data privacy, and multi-omics integration.
• Future directions involve multimodal AI, explainable AI (XAI), federated learning, precision nutrition, and intelligent devices.
• AI offers a roadmap for personalized health optimization beyond population-averaged guidelines.