Toward a species-based approach to One Health by leveraging large language models

One Health is conceptually framed as an integrated approach linking human, animal, plant, and ecosystem health. In practice, however, the ecosystem dimension—including biodiversity processes and dynamics—remains comparatively underdeveloped. At the same time, biodiversity is increasingly recognized as shaping health through ecological and socio-cultural pathways mediated by individual species and species communities. Yet, knowledge about links between biodiversity and health is fragmented across taxa, regions, and disciplines. This fragmentation limits synthesis, reuse, and comparison, and slows the integration of biodiversity into One Health research and policy. To address this gap, we propose a species-based approach to consolidate biodiversity–health knowledge within One Health. Specifically, we developed a framework and evaluated its feasibility for extracting and structuring Species Contributions to People (SCP) by leveraging large language models (LLMs). The framework combines corpus preparation, semantic extraction, validation, and dataset assembly. Together, these steps are designed to transform fragmented textual evidence into standardized, species-level records linked to defined health outcomes. To demonstrate its use, we implemented the framework as an R workflow and applied it in a proof-of-concept case study focused on vector-borne and zoonotic diseases in the Amazon region. Our results showed that species–health links can be extracted and structured into analysis-ready data. We further demonstrated how curated evidence and evaluation sets enable measurement of extraction performance and support targeted expert review. Overall, this proof-of-concept application demonstrates the feasibility of using LLM-assisted workflows to consolidate species–health evidence into structured SCP records. It also provides a foundation for building an extensible SCP data resource. With expanded coverage, systematic benchmarking, and community governance, such a resource could support the identification of nature-based solutions for health and advance the integration of biodiversity within the ecosystem dimension of One Health.

Authors

Institutions

Publication Details

Journal
Next Sustainability
Published
2026-11-14
DOI
https://doi.org/10.1016/j.nxsust.2026.100590
Primary Topic
Zoonotic diseases and public health
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Toward a species-based approach to One Health by leveraging large language models

Antoine Adde, Florian Altermatt, François Keck
2 links
Next Sustainability
Zoonotic diseases and public health
article

Toward a species-based approach to One Health by leveraging large language models

Antoine Adde, Florian Altermatt, François Keck
article en

Abstract

One Health is conceptually framed as an integrated approach linking human, animal, plant, and ecosystem health. In practice, however, the ecosystem dimension—including biodiversity processes and dynamics—remains comparatively underdeveloped. At the same time, biodiversity is increasingly recognized as shaping health through ecological and socio-cultural pathways mediated by individual species and species communities. Yet, knowledge about links between biodiversity and health is fragmented across taxa, regions, and disciplines. This fragmentation limits synthesis, reuse, and comparison, and slows the integration of biodiversity into One Health research and policy. To address this gap, we propose a species-based approach to consolidate biodiversity–health knowledge within One Health. Specifically, we developed a framework and evaluated its feasibility for extracting and structuring Species Contributions to People (SCP) by leveraging large language models (LLMs). The framework combines corpus preparation, semantic extraction, validation, and dataset assembly. Together, these steps are designed to transform fragmented textual evidence into standardized, species-level records linked to defined health outcomes. To demonstrate its use, we implemented the framework as an R workflow and applied it in a proof-of-concept case study focused on vector-borne and zoonotic diseases in the Amazon region. Our results showed that species–health links can be extracted and structured into analysis-ready data. We further demonstrated how curated evidence and evaluation sets enable measurement of extraction performance and support targeted expert review. Overall, this proof-of-concept application demonstrates the feasibility of using LLM-assisted workflows to consolidate species–health evidence into structured SCP records. It also provides a foundation for building an extensible SCP data resource. With expanded coverage, systematic benchmarking, and community governance, such a resource could support the identification of nature-based solutions for health and advance the integration of biodiversity within the ecosystem dimension of One Health.

Next SustainabilityVol. 8
University of Zurich (CH), Swiss Federal Institute of Aquatic Science and Technology (CH)
Board of the Swiss Federal Institutes of Technology
Zoonotic diseases and public health
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon