Artificial Intelligence and Precision Nutrition

WHAT IS PRECISION NUTRITION?

Precision nutrition is used to evaluate various factors of our lifestyle and create nutritional approaches to manage chronic diseases and improve health (¹). Preventing chronic disease will lead to better quality of life and tremendous savings in health care costs. Zeisel (²) states that individuals vary in their reactions to nutrients in the diet due to “metabolic heterogeneity” (e.g., differences in things like genetics, lifestyle, and the environment). Precision nutrition is not meant to lead to individualized dietary prescriptions, but rather to classify people into similar groups to help better identify their dietary needs (²).

WHAT IS ARTIFICIAL INTELLIGENCE?

Artificial intelligence (AI) is an everyday part of our daily lives, although many of us do not recognize it. For example, AI is used in weather forecasting, facial recognition, natural language processing, collaborative recommendations, industrial processing improvements, and financial analyses (³). AI algorithms might help to predict the connections between nutrition and health, leading to improved dietary assessment, especially with respect to self-reporting errors (⁴). In addition, AI is used to extract large amounts of data that can help to assess dietary behaviors (⁴).

Machine learning is a type of AI, and it can be altered after exposure to novel data sets and “learn” without new programming (³). In addition, deep learning has been involved in the prediction of certain diseases (^5,6).

WHY IS AI IMPORTANT WITH RESPECT TO PRECISION NUTRITION?

Schwalbe and Wahl (⁷) state that “AI-driven health interventions fit into four categories relevant to global health researchers: 1) diagnosis, 2) patient morbidity or mortality risk assessment, 3) disease outbreak prediction and surveillance, and 4) health policy and planning.”

Morgenstern et al. (⁸) discussed how improvements in big data and machine learning can address measurement error in the field of nutritional epidemiology. For example, they discussed how traditional dietary collection methods might not allow for the proper amount of data. Using machine learning could lead to improving the precision and validity of dietary intake assessments. “Overall, we propose that judicious application of big data and machine learning in nutrition science could offer new means of dietary measurement, more tools to model the complexity of diet and its relations with diseases, and additional potential ways of addressing confounding” (⁹).

For example, Celis-Morales et al. (¹⁰) conducted a randomized controlled trial where they evaluated personalized nutrition approaches from using an online tool that provides nutritional advice. Of the 1,269 volunteers who completed their study, those who were provided personalized nutrition programs improved their diets, indicated by higher Health Eating Index scores.

Berry et al. (¹¹) included 1,002 twins and healthy adults who were unrelated to the twins in the United Kingdom in the PREDICT 1 study. They evaluated metabolic responses after a meal, both in the clinic and at home. They reported considerable interindividual variability in postprandial measures of blood triglyceride, glucose, and insulin concentrations after duplicate meals were given. In addition, they reported that variables like the gut microbiome had a greater effect (explained 7.1% of the variance) on postprandial triglyceride and glucose concentrations than the macronutrient (carbohydrate, fat, and protein) composition of the meal. They developed a machine learning model that predicted both triglyceride and glucose responses to nutrition intake (personalized nutrition).

SUMMARY

AI and machine learning technology are an expanding part of our daily lives. Because so much of our lives already include AI, it would benefit health professionals to learn more about this technology and how it could be advantageous to their clients. Although additional research studies using big data sets are required to better elucidate how much AI and/or machine learning affects precision nutrition, the current focus on improving individual lifestyle and associated nutritional intake should continue (¹²).