“Nutritional Epidemiology × Data Science and AI”: Creating New Value in the Fields of Food and Health, and Solving Real-world Challenges

Abstract

Due to its complexity, the epidemiological approach for clarifying the “why” at the heart of nutritional epidemiology remains poorly understood. Furthermore, as data science and AI, which underlie the methods employed for addressing the “how” of problem-solving, become more widespread, it is becoming increasingly clear that there is a lack of educational opportunity for gaining a comprehensive and deep understanding of both theory and technology. This review examines the prospects of “nutritional epidemiology integrated with data science and AI” in the Society 5.0 era. It introduces key terms from recent epidemiological research focused on causal inference and provides an overview of how these theories can be applied in practice. By doing so, it highlights the limitations and challenges of nutritional epidemiology when using data science and AI, while clearly explaining the key factors needed for translation of research findings into practical applications in public health and clinical settings. Furthermore, it highlights the importance of creating new value through AI analysis of big data and data-driven real-world data, and outlines the key principles for translating evidence-based findings in medicine and nutrition into the formulation and societal implementation of health policies and dietary guidelines.