Mapping disease-environment connections reported in scientific literature at scale with EasyNER
(2025) EurIPS 2025- Abstract
- Climate change and connected environmental disruptions have a large negative
effect on the health of humans and other species. Improved understanding of
known links in the environment-disease nexus as well as knowledge gaps is
essential to prepare for and mitigate this escalating threat. However, knowledge
fragments are scattered across millions of scientific articles making information
synthesis too complex even for large review consortia. Natural language
processing (NLP) tools could help execute this at scale. Here, we developed an
information synthesis pipeline for mapping links between environmental
processes and diseases which expanded our EasyNER tool. We then applied it to
extract co-mentions... (More) - Climate change and connected environmental disruptions have a large negative
effect on the health of humans and other species. Improved understanding of
known links in the environment-disease nexus as well as knowledge gaps is
essential to prepare for and mitigate this escalating threat. However, knowledge
fragments are scattered across millions of scientific articles making information
synthesis too complex even for large review consortia. Natural language
processing (NLP) tools could help execute this at scale. Here, we developed an
information synthesis pipeline for mapping links between environmental
processes and diseases which expanded our EasyNER tool. We then applied it to
extract co-mentions from 16.7 million abstracts in PubMed, a large collection of
life science literature, producing a knowledge graph of the environment-disease
nexus. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/7e44ba10-56ae-4b56-bf9e-d2b61a1eeecb
- author
- Aasa, Carl
; Boyd, Emily
LU
and Aits, Sonja
LU
- organization
-
- Cell Death, Lysosomes and Artificial Intelligence (research group)
- BECC: Biodiversity and Ecosystem services in a Changing Climate
- LU Profile Area: Proactive Ageing
- LU Profile Area: Nature-based future solutions
- LU Profile Area: Natural and Artificial Cognition
- LTH Profile Area: Engineering Health
- LTH Profile Area: AI and Digitalization
- EpiHealth: Epidemiology for Health
- LUCC: Lund University Cancer Centre
- eSSENCE: The e-Science Collaboration
- LUCSUS (Lund University Centre for Sustainability Studies)
- publishing date
- 2025-12-02
- type
- Contribution to conference
- publication status
- published
- subject
- keywords
- natural language processing, Named entity recognition, environmental health, environmental toxicology, Biodiversity, Climate change, Artificial intelligence, Deep learning, Machine learning, software, BioBERT
- conference name
- EurIPS 2025
- conference location
- Copenhagen, Denmark
- conference dates
- 2025-12-02 - 2025-12-07
- project
- Biomedical text mining for systems biology
- The Swedish Centre of Excellence on Impacts of Climate Extremes (Climes)
- Artificial intelligence-based text mining for COVID-19 and other areas of medicine
- Environmental toxins as cause of lysosomal damage-induced cell death in animals and humans
- Lund University AI Research
- Revealing drivers of cell death disruption across species and theri links to biodiversity loss and human disease
- Lysosomes in cell death - from molecular mechanisms to new treatment strategies
- language
- English
- LU publication?
- yes
- id
- 7e44ba10-56ae-4b56-bf9e-d2b61a1eeecb
- alternative location
- https://github.com/climateainordics/AICC2025/blob/main/papers/Mapping_Aasa.pdf
- date added to LUP
- 2026-01-15 08:32:55
- date last changed
- 2026-01-15 08:46:28
@misc{7e44ba10-56ae-4b56-bf9e-d2b61a1eeecb,
abstract = {{Climate change and connected environmental disruptions have a large negative<br/>effect on the health of humans and other species. Improved understanding of<br/>known links in the environment-disease nexus as well as knowledge gaps is<br/>essential to prepare for and mitigate this escalating threat. However, knowledge<br/>fragments are scattered across millions of scientific articles making information<br/>synthesis too complex even for large review consortia. Natural language<br/>processing (NLP) tools could help execute this at scale. Here, we developed an<br/>information synthesis pipeline for mapping links between environmental<br/>processes and diseases which expanded our EasyNER tool. We then applied it to<br/>extract co-mentions from 16.7 million abstracts in PubMed, a large collection of<br/>life science literature, producing a knowledge graph of the environment-disease<br/>nexus.}},
author = {{Aasa, Carl and Boyd, Emily and Aits, Sonja}},
keywords = {{natural language processing; Named entity recognition; environmental health; environmental toxicology; Biodiversity; Climate change; Artificial intelligence; Deep learning; Machine learning; software; BioBERT}},
language = {{eng}},
month = {{12}},
title = {{Mapping disease-environment connections reported in scientific literature at scale with EasyNER}},
url = {{https://github.com/climateainordics/AICC2025/blob/main/papers/Mapping_Aasa.pdf}},
year = {{2025}},
}