Identifying Biomedical Entities for Datasets in Scientific Articles: 4-Step Cache-Augmented Generation Approach Using GPT-4o and PubTator 3.0
This study highlights the potential of large language model-supported metadata annotation. The findings underscore the practical feasibility of full-text analysis and suggest its potential for integration into routine workflows for biomedical metadata generation.
Authors
- Harald Binder (ORCID: https://orcid.org/0000-0002-5666-8662)
- Olaf Groß (ORCID: https://orcid.org/0000-0001-8660-3619)
- Gita Benadi (ORCID: https://orcid.org/0000-0002-5263-422X)
- Guido Schwarzer (ORCID: https://orcid.org/0000-0001-6214-9087)
- Robert Zeiser (ORCID: https://orcid.org/0000-0001-6565-3393)
- Manuel Watter (ORCID: https://orcid.org/0000-0001-7831-0613)
- Klaus Kaier (ORCID: https://orcid.org/0000-0003-0837-6945)
- Felix Engel (ORCID: https://orcid.org/0009-0008-3871-9762)
- Claudia Giuliani (ORCID: https://orcid.org/0009-0004-1172-3244)
- Jonas Werner (ORCID: https://orcid.org/0009-0009-0891-7503)
Institutions
- University of Freiburg (DE)
- University Medical Center Freiburg (DE)
Publication Details
- Journal
- JMIR Formative Research
- Published
- 2025-11-20
- DOI
- https://doi.org/10.2196/73822
- Citations
- 3
- Primary Topic
- Biomedical Text Mining and Ontologies
- Type
- article
- Field-Weighted Citation Impact
- 2.42