Session : Artificial Intelligence for Personalized Nutrition

In this presentation, we explore the role of Artificial Intelligence (AI) in global nutrition. We will begin by defining AI, with a focus on machine learning, and contextualize nutrition within the broader framework of global health. Firstly, this talk will delve into various AI strategies and their relevance to nutrition. We will carefully examine national strategies that place special emphasis on AI and its application in the field of nutrition. Next, our presentation will concentrate on concrete use cases in nutrition. We will illustrate diverse examples of AI applications, such as climate change forecasting for agriculture, logistics, and fresh produce management, as well as its implications in health, particularly regarding microbiota, fertility, obesity, and public health, among others. With a global approach in mind, we will shed light on the complex relationship between humans, nature, and artificial intelligence. We will assess the technological advancements in AI while addressing the inequalities that may arise between countries in this context. We will underscore the importance of interdisciplinarity for a balanced approach. This involves both respecting the environment and biodiversity, as well as involving local communities in the design and implementation of these new technologies. In conclusion, we will closely examine the issue of education in an interdisciplinary context, presenting activities between Laval University and Sorbonne University as an example. This will demonstrate how interdisciplinarity can foster innovative and sustainable solutions to meet the challenges related to AI and nutrition on a global scale.

Concrete examples of pharmaco-nutrition research projects leveraging AI will be presented.

Hyperandrogenism, which refers to the excessive production of male hormones in women, affects 5 to 10% of women. Its diagnosis in children or adolescents requires invasive blood tests, and their reliability is not optimal. We will discuss how the use of AI has recently led to the development of new diagnostic tools.

I will present some recent work and where AI may have the most significant impact on Nutrition research. Monitoring metabolites is one of the primary means of better understanding our health and the impact of nutritional intake on our well-being.

Bacteria are the most studied organisms in the gut microbiome as they are recognized to have a profound impact on human gut health. The relationship between the gut bacteria and their host is often described as symbiotic, as the bacteria enjoy a constant source of nutrients and a stable environment, while the host benefits from additional metabolic capabilities and protection against pathogens. Albeit being the most abundant biological entities on the planet, the viruses that infect them, called bacteriophages or phages, have received much less attention. Recent studies have shown that phages modulate microbiome community dynamics due to their obligate predator-prey interactions with their bacterial hosts. Thus, bacteriophages are also increasingly believed to play a role in human health. A study performed on 116 healthy adult Quebecers explored the interplay between the bacterial fraction of the gut microbiota, the endocannabinoid system and human health. Importantly, study subjects were chosen to represent the general population with different body compositions, dietary habits, and levels of daily physical activity. Individuals were recruited at the Institute of Nutrition and Functional Foods at Université Laval and the anthropometric and metabolic characteristics of all subjects were measured. In addition, whole metagenome shotgun sequencing was performed on fecal samples from these individuals. For this ongoing project, we are exploring the viral component from this sequencing data. Artificial intelligence (AI) methods will be applied to investigate the associations between phages, bacteria, and human health. Particular attention will be placed upon choosing interpretable approaches. AI models will be trained on microbial abundance data to predict variables of metabolic and anthropometric health. Then, these models will be studied to identify phages and bacteria biomarkers of those health variables. An emphasis will be put on understanding the interactions between phages and bacteria. These analyses will further our understanding of the roles that phages and phage-bacteria interactions play in nutrition and health.