Leveraging large language models to populate structured clinical case report forms from unstructured medical notes in radiation oncology
This study demonstrated the effectiveness of LLMs in structuring clinical data from medical non-English notes, with high accuracy in extracting and categorizing information. While multi-institutional validation is needed, the results indicate a significant healthcare impact through efficient data management, processing notes in 16 s, and accurately populating CRFs with minimal staff involvement.
Authors
- Marcel Büttner
- Marcel Nachbar (ORCID: https://orcid.org/0000-0001-8923-4740)
- Augusto García-Agúndez (ORCID: https://orcid.org/0000-0002-5440-1032)
- Daniela Thorwarth (ORCID: https://orcid.org/0000-0003-1433-9315)
- Maximilian Niyazi
- Carsten Eickhoff
- Cihan Gani
- Nianzi Yi
Institutions
- German Cancer Research Center (DE)
- University of California, San Francisco (US)
- Heidelberg University (DE)
- Innovation Cluster (Canada) (CA)
- University Children's Hospital Tübingen (DE)
- University of Tübingen (DE)
Publication Details
- Journal
- Clinical and Translational Radiation Oncology
- Published
- 2026-03-09
- DOI
- https://doi.org/10.1016/j.ctro.2026.101143
- Primary Topic
- Topic Modeling
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Deutsche Forschungsgemeinschaft