Methodological Advances of AI and Business Intelligence in Nutrition: Techniques, Applications, and Future Directions

Abstract

Artificial Intelligence (AI) and Business Intelligence (BI) are increasingly transforming nutritional science by enabling precise dietary assessment, personalized nutrition planning, and data-driven health monitoring. This review provides an integrated analysis of methodological advancements in AI and BI techniques applied within nutrition research. A structured search was conducted across PubMed, Scopus, Web of Science, Google Scholar, and ScienceDirect. After removing duplicates and applying predefined inclusion and exclusion criteria—focusing on studies that utilized AI, machine learning, deep learning, or BI frameworks for dietary assessment, nutrient estimation, predictive modeling, or personalized dietary recommendations—a total of 73 studies met the eligibility criteria and were included in the final synthesis. The review categorizes methodological approaches, highlights their strengths and limitations, and evaluates their practical implications for clinical and public-health nutrition. While AI holds significant promise for improving accuracy, scalability, and personalization in nutrition, several challenges remain, including dataset limitations, model interpretability, and ethical considerations. The findings emphasize the need for culturally diverse datasets, explainable models, and integrated AI–BI architectures to advance future research and real-world implementation. Integrating AI and nutrition, it still faces several data-related, methodological, and ethical limitations