AI in Nutrition Research: Forging Innovations for Global Nutrition and Health Surveillance

After several decades of foundational research on artificial intelligence (AI), machine learning (ML), and data science, the landscape of knowledge and information generation and dissemination has evolved dramatically. Buzzwords like big data, deep learning, augmented reality, and chatbots have taken center stage and the recent emergence of free tools, capable of generating text, sound, and images has brought forth a new wave of technical, ethical, and existential questions. In nutrition research and practice, global nutrition surveillance systems have started to play a critical role in national health monitoring by providing the data necessary to understand, prevent, and address nutritional issues in a population. The national and global nutrition surveillance systems are expanding their reach and improving data quality and standardization. We assessed several major nutrition and food data dashboards developed and maintained by prominent academic and intergovernmental organizations. To ensure systematic and comprehensive evaluation, we developed and tested four key principles: Evidence, Efficiency, Emphasis, and Ethics, or 4E principles, and proposed 48 quality evaluation metrics for nutrition and food dashboards. We determined the major challenges of modern dashboards related to data granularity, sharing, visualization, and literacy. To address some of those challenges, we designed the Global Nutrition and Health Atlas (GNHA: https://sites.tufts.edu/gnha/), an open-access online platform covering nutrition and health data with 26 themes and 500+ indicators from 190+ countries over the last 30 years. We view GNHA as an interactive tool aiming to share information and perspectives and foster collaborations and innovations. As the next step of the platform development, we focus on incorporating interactive analytical tools, high-quality data visualization, standardization of data capture, and educational material, such as online guides and tutorials. The merging of nutrition surveillance and AI could contribute substantially to better public health outcomes, improved resource allocation, and discovery acceleration.