Speaker
Description
The concept of planetary health acknowledges the complex relationships between human health and the environment, emphasizing the need for sustainable stewardship of natural systems. To integrate this perspective into policy and decision-making, the Planetary Health Impact Assessment (PHIA) has been proposed, an adaption of the Health Impact Assessment (HIA) to planetary-scale challenges. Developing PHIA framework requires a methodology to summarise and synthesize knowledge on how human-driven environmental changes impact human health, both directly and indirectly. This study explores the development of an initial PHIA framework using a knowledge graph (KG), with RF-EMF exposure as a case study.
A KG was constructed through workshops with 13 experts from diverse disciplines, mapping possible pathways linking RF-EMF exposure to health effects on various organisms, including humans, and potential indirect effects on humans through ecological consequences. Additionally, two AI tools—a Natural Language Processing (NLP)-based automated extraction tool and OpenAI’s ChatGPT—were tested for their ability to extract the same information and generate supplementary KGs from scientific literature.
The findings highlight the expert-based KG as a consensus-driven, adaptable way for summarising and communicating multidisciplinary knowledge and guiding research. In this context, AI tools, while efficient in literature processing, require human oversight to ensure accuracy and contextual understanding. This study marks an initial step toward transitioning to PHIA frameworks into research and policymaking. Future advancements in AI are necessary to enhance their ability to contribute independently.
