A Lightweight Framework for Medical Named Entity Recognition and Relation Extraction
Studies in Health Technology and Informatics · 2026
DOI: 10.3233/shti261002 ↗Kurz gefasst
Fachartikel in der Zeitschrift Studies in Health Technology and Informatics, 2026. Verfasst von Sihan Wu und 3 weiteren Autor:innen. Beteiligte Einrichtung: Technical University of Munich, TUM School of Medicine and Health, Institute for AI and Informatics in Medicine (AIIM), TUM University Hospital. Hauptthemen laut MEDLINE: Electronic Health Records, Neural Networks, Computer, Natural Language Processing, Information Storage and Retrieval. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
Introduction
Clinical text in electronic health records contains valuable information for medical research. Named entity recognition (NER) and relation extraction (RE) are key tasks for structuring this information, but existing deep learning approaches are often computationally expensive.
Methods
We develop a lightweight, pipeline-based NER–RE system using neural networks with contextualized embeddings based on spaCy and Thinc. The impact of data preprocessing and hyperparameter tuning (HPT) on NER is evaluated. RE is performed using a separate classifier on candidate entity pairs. Experiments are conducted on i2b2 and n2c2 datasets.
Results
The approach achieves strong NER performance after optimization. Better data quality or applying HPT improve performance comparably, but combining both yields only marginal gains additionally.
Discussion
Lightweight models are effective for NER but limited for RE due to error propagation. Our NER results also suggest that HPT can compensate for data quality issues.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Technical University of Munich, TUM School of Medicine and Health, Institute for AI and Informatics in Medicine (AIIM), TUM University Hospital.
Zitieren
APA
Wu, S., Gordon, S., Boeker, M., & Modersohn, L. (2026). A Lightweight Framework for Medical Named Entity Recognition and Relation Extraction. Studies in Health Technology and Informatics. https://doi.org/10.3233/shti261002
Vancouver
Wu S, Gordon S, Boeker M, Modersohn L. A Lightweight Framework for Medical Named Entity Recognition and Relation Extraction. Studies in Health Technology and Informatics. 2026. doi:10.3233/shti261002
BibTeX
@article{Wu2026,
author = {Wu, Sihan and Gordon, Susanna and Boeker, Martin and Modersohn, Luise},
title = {A Lightweight Framework for Medical Named Entity Recognition and Relation Extraction},
journal = {Studies in Health Technology and Informatics},
year = {2026},
doi = {10.3233/shti261002}
}
EUVIMED