Developing a natural language processing system using transformer-based models for adverse drug event detection in electronic health records
PLOS One · 01.07.2026 · 21 (7) · e0350516
DOI: 10.1371/journal.pone.0350516 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS One, online erschienen am 01.07.2026. Verfasst von Jingyuan Wu und 4 weiteren Autor:innen. Beteiligte Einrichtung: Data Science Institute, Vanderbilt University, Nashville, Tennessee, United States of America. Hauptthemen laut MEDLINE: Natural Language Processing, Electronic Health Records, Drug-Related Side Effects and Adverse Reactions. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Institute of General Medical Sciences, the National Institute of Child Health and Human Development.
Abstract (Originalsprache)
Objective
To develop a transformer-based natural language processing (NLP) system for detecting adverse drug events (ADEs) from clinical notes in electronic health records (EHRs). Materials and Methods We fine-tuned BERT Short-Formers and Clinical-Longformer using the processed dataset from the 2018 National NLP Clinical Challenges (n2c2) shared task Track 2. Two data processing methods, window-based and split-based approaches, were compared to identify the optimal processing method. Model generalizability was evaluated on a dataset extracted from Vanderbilt University Medical Center (VUMC) EHRs.
Results
On the n2c2 dataset, the best 5-fold cross-validation AUPRC, micro F1, and macro F1 scores were 0.840 (Clinical-Longformer, 4-chunk split), 0.965 (BioBERT, 15-word window), and 0.852 (Clinical-Longformer, 10-chunk split). On the VUMC dataset, the best AUPRC, micro F1, and macro F1 scores were 0.536 (Clinical-Longformer, 6-chunk split), 0.966 (BERT-base-uncased, 6-chunk split), and 0.762 (Clinical-Longformer, 4-chunk split).
Discussion
Transformer-based models demonstrated strong performance for ADE detection, with split-based processing generally outperforming window-based methods. Clinical-Longformer combined with a practical split-based approach showed promise for real-world implementation. Beyond token limits, chunk size substantially influenced model performance, even when text length remained within limits.
Conclusion
Our findings provide practical guidance for developing transformer-based ADE detection systems from clinical notes. The selection of both text preprocessing strategies and model architectures should be guided by note characteristics and practical considerations such as annotation burden.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Data Science Institute, Vanderbilt University, Nashville, Tennessee, United States of America.
- Department of Psychology and Human Development, Vanderbilt University, Nashville, Tennessee, United States of America.
- Department of Biostatistics, Vanderbilt University Medical Center, Nashville, Tennessee, United States of America.
- Department of Pediatrics, Vanderbilt University Medical Center, Nashville, Tennessee, United States of America.
ORCID-Profile: Jingyuan Wu, Leena Choi
Förderung
- National Institute of General Medical Sciences
- the National Institute of Child Health and Human Development
Zitieren
APA
Wu, J., Ruan, X., McNeer, E., Rossow, K. M., & Choi, L. (2026). Developing a natural language processing system using transformer-based models for adverse drug event detection in electronic health records. PLOS One, 21(7), e0350516. https://doi.org/10.1371/journal.pone.0350516
Vancouver
Wu J, Ruan X, McNeer E, Rossow KM, Choi L. Developing a natural language processing system using transformer-based models for adverse drug event detection in electronic health records. PLOS One. 2026;21(7):e0350516. doi:10.1371/journal.pone.0350516
BibTeX
@article{Wu2026,
author = {Wu, Jingyuan and Ruan, Xiaodi and McNeer, Elizabeth and Rossow, Katelyn M and Choi, Leena},
title = {Developing a natural language processing system using transformer-based models for adverse drug event detection in electronic health records},
journal = {PLOS One},
year = {2026},
volume = {21},
number = {7},
pages = {e0350516},
doi = {10.1371/journal.pone.0350516}
}
EUVIMED