Large language model-based screening of substances and their composition from safety data sheets for high-resolution chemical exposure assessment
- 1School of Public Health, Seoul National University, Seoul, Republic of Korea
- 2Environmental Risk Research Division, National Institute of Environment Research, Incheon, Republic of Korea
- 3UCD Centre for Safety and Health at Work, School of Public Health, Physiotherapy, and Sports Science, University College Dublin, Dublin, Ireland
Abstract
Background: SDSs provide information on chemical substances in professional and consumer products. Large language models (LLMs) offer a rapid approach to screening chemical information from SDSs.
Objective: This study aimed to validate the performance of LLMs in accurately extracting substance information from SDSs of products.
Methods: Chemical information was extracted from the SDSs of cleaning products using the LLMs ChatGPT-4o and Gemini 2.5 Pro. The performance of the LLMs was evaluated against manually extracted data using precision, sensitivity, and F1 score.
Results: A total of 301 substance–composition combinations across 59 products were included in the validation. The Gemini 2.5 Pro model showed a higher F1 score (1.00) than ChatGPT-4o (0.94).
Significance: LLMs enable high-throughput chemical information extraction from SDSs, reducing the burden of manual screening and supporting large-scale combined-exposure assessments.
Impact Statement: The accurate identification of chemical substances in professional and consumer products is a major challenge in assessing complex chemical exposures in exposure science and epidemiology. This study validated the use of Large Language Models (LLMs) as a novel, rapid, and highly accurate method of extracting high-throughput chemical data from multilingual Safety Data Sheets. Our findings illustrated that LLMs could be effectively used to overcome the labor-intensive and time-consuming limitations of manual data screening. This approach allows high-throughput analysis, enabling comprehensive combined-exposure assessments in large-scale epidemiological and exposure assessment studies. Additionally, as LLMs can extract data in different languages, they can facilitate international research collaboration.
Related
- Ultra-High-Performance Liquid Chromatography-High Resolution Mass Spectrometry and Gas Chromatography-Mass Spectrometry Screening of Classical Psychoactive Substances, Novel Psychoactive Substances, and Metabolites in Postmortem Urine Samples
- Assessing the use of ASAP‐MS for Screening a UK Pub for Illicit Drug Substances
- Capillary electrophoresis coupled with molecularly imprinted polymer solid phase extraction for screening malachite green and related substances
- Auditing frozen chemical language model embeddings for drug-delivery screening: fingerprint baselines, random-initialization controls, and scaffold-memorization control tasks
- Assessment of Antioxidant Capacity, Phytochemical Phenolic Composition, Cholinesterase and α-Glucosidase Inhibitory Activities, and Antimicrobial Potential of Tribulus terrestris
BibTeX
@article{lee2026large,
title = {{Large language model-based screening of substances and their composition from safety data sheets for high-resolution chemical exposure assessment}},
author = {Daeyeop Lee and Kiyoung Lee and Sewon Lee},
journal = {Journal of Exposure Science \& Environmental Epidemiology},
year = {2026},
month = may,
doi = {10.1038/s41370-026-00917-z},
url = {https://doi.org/10.1038/s41370-026-00917-z},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Large language model-based screening of substances and their composition from safety data sheets for high-resolution chemical exposure assessment
AU - Daeyeop Lee
AU - Kiyoung Lee
AU - Sewon Lee
PY - 2026
DA - 2026/05/13
JO - Journal of Exposure Science & Environmental Epidemiology
DO - 10.1038/s41370-026-00917-z
UR - https://doi.org/10.1038/s41370-026-00917-z
AB - Background: SDSs provide information on chemical substances in professional and consumer products. Large language models (LLMs) offer a rapid approach to screening chemical information from SDSs. Objective: This study aimed to validate the performance of LLMs in accurately extracting substance information from SDSs of products. Methods: Chemical information was extracted from the SDSs of cleaning products using the LLMs ChatGPT-4o and Gemini 2.5 Pro. The performance of the LLMs was evaluated against manually extracted data using precision, sensitivity, and F1 score. Results: A total of 301 substance–composition combinations across 59 products were included in the validation. The Gemini 2.5 Pro model showed a higher F1 score (1.00) than ChatGPT-4o (0.94). Significance: LLMs enable high-throughput chemical information extraction from SDSs, reducing the burden of manual screening and supporting large-scale combined-exposure assessments. Impact Statement: The accurate identification of chemical substances in professional and consumer products is a major challenge in assessing complex chemical exposures in exposure science and epidemiology. This study validated the use of Large Language Models (LLMs) as a novel, rapid, and highly accurate method of extracting high-throughput chemical data from multilingual Safety Data Sheets. Our findings illustrated that LLMs could be effectively used to overcome the labor-intensive and time-consuming limitations of manual data screening. This approach allows high-throughput analysis, enabling comprehensive combined-exposure assessments in large-scale epidemiological and exposure assessment studies. Additionally, as LLMs can extract data in different languages, they can facilitate international research collaboration.
SN - 1559-0631
SN - 1559-064X
ER -