AI Revolutionizing Precision Nutrition: A Comprehensive Review

Precision nutrition is an advanced approach to dietary planning that tailors nutritional guidance to individual characteristics, including genetics, lifestyle, and environmental factors. The integration of artificial intelligence (AI) into precision nutrition opens up unprecedented opportunities to enhance the efficacy and personalization of nutritional recommendations. AI can analyze vast amounts of data from diverse sources, such as multiomic profiles, dietary habits, and medical histories, enabling the identification of nuanced dietary needs at the individual level. This review explores the latest advancements in AI and their applications in precision nutrition.

A scoping review strategy following the Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) was utilized. Inclusion criteria included articles ranging from December 8, 2014, to May 28, 2024, in English sourced from reputable academic databases. Search keywords were categorized into precision nutrition, AI, and natural language processing. Exclusion criteria included editorials, errata, letters, notes, comments, and animal studies. A total of 198 relevant articles were included in this literature review.

The 198 articles were disseminated across 142 venues, comprising 98 journals and 44 conferences. The journals were manually classified into five distinct categories: Clinical Medicine, Food & Nutrition Science, Informatics, Computer Science, and Biology. This distribution reflects a high level of interest and activity in Clinical Medicine and Informatics, suggesting a strong focus on applying AI techniques in clinical settings for personalized nutrition interventions or medical applications.

Among the 198 publications analyzed, 99 specifically studied one or more diseases. The top three most studied diseases are diabetes, cardiovascular diseases, and cancers. Less studied diseases include gastrointestinal disorders, neurodegenerative diseases, eating disorders, mental health disorders, obesity, eye fatigue, COVID-19, food allergies, and skin disease. Research on these less-studied diseases mostly emerged after 2020.

The main applications of AI in precision nutrition are health optimization, disease prevention, and disease management. Health optimization aims to enhance individuals’ well-being through personalized nutrition interventions using various AI approaches. Disease prevention focuses on using AI to predict and prevent the onset of diseases through personalized dietary recommendations. Disease management involves using AI to assist in the management of existing diseases through tailored nutritional plans.

The review highlights diverse datasets used in the literature and summarizes methodologies and evaluation metrics to guide future studies. AI methods are systematically categorized, with each method described alongside examples from precision nutrition research. Evaluation metrics used to assess AI models are also categorized and explained with relevant examples.

The review emphasizes the importance of minority and cultural perspectives in promoting equity for precision nutrition using AI. It explores the impact and potential of various factors, such as socioeconomics, cultural sensitivity, technology accessibility and digital literacy, ethical and privacy concerns, personalized nutrition needs, community-based approaches, and policy and advocacy, on AI for precision nutrition.

Future research should further integrate minority and cultural factors to fully harness AI’s potential in precision nutrition. There is a need for more comprehensive research, detailed information about research methods, and research materials, including detailed dataset links and descriptions. Addressing the fragmented knowledge and scattered publication patterns is also crucial.

This scoping review provides a comprehensive overview of the current landscape of AI in precision nutrition, highlighting the advancements, challenges, and future directions. By examining publication venues, targeted diseases, applications, datasets, AI methods, evaluation metrics, and minority and cultural factors, this review improves the understanding of AI’s potential in precision nutrition and provides new directions for future research.

AD: Alzheimer’s disease, AI: artificial intelligence, ANOVA: analysis of variance, AUC: area under the curve, AUROC: area under the receiver operating characteristic, CGM: continuous glucose monitoring, CRC: colorectal cancer, DSS: decision support system, EHR: electronic health record, EN: enteral nutrition, FEL: food exchange list, FFQ: Food Frequency Questionnaire, HbA1c: hemoglobin A1c, HEI: Healthy Eating Index, ICU: intensive care unit, LLM: large language model, LSTM: long short-term memory, MIMIC-IV: Medical Information Mart for Intensive Care IV, NLP: natural language processing, PPGR: postprandial glycemic response