Comparative Evaluation of Proprietary and Open-Source Large Language Models for Systematic Multi-source Information Extraction in Interventional Oncology
Proprietary LLMs can accurately extract most key TACE-related variables from routine clinical reports and may support decision-making in interventional oncology; however, all models showed poor performance in vascular invasion detection and follow-up assessment, so expert human oversight remains essential.
Authors
- Keno K. Bressem (ORCID: https://orcid.org/0000-0001-9249-8624)
- Elif Can (ORCID: https://orcid.org/0000-0001-5319-7570)
- Sebastian Gassenmaier (ORCID: https://orcid.org/0000-0003-3701-2227)
- Wibke Uller (ORCID: https://orcid.org/0000-0001-6756-9005)
- Felix Busch (ORCID: https://orcid.org/0000-0001-9770-8555)
- Michael P. Brönnimann (ORCID: https://orcid.org/0000-0003-0552-4850)
- Marcus R. Makowski (ORCID: https://orcid.org/0000-0001-8778-647X)
- Saif Afat (ORCID: https://orcid.org/0000-0002-6861-6245)
- Lisa C Adams (ORCID: https://orcid.org/0000-0002-7449-9428)
- Avan Kader (ORCID: https://orcid.org/0000-0002-5453-1267)
- Michael Doppler
- Katharina Sophie Vogt (ORCID: https://orcid.org/0000-0003-3911-637X)
- Aboelyazid Elkilany (ORCID: https://orcid.org/0000-0002-4311-0520)
- Elmar Kotter (ORCID: https://orcid.org/0000-0001-9022-6000)
- Raid Alshinibr
Institutions
- University of Bern (CH)
- University Hospital of Bern (CH)
- University Medical Center Freiburg (DE)
- TUM Klinikum (DE)
- Humboldt-Universität zu Berlin (DE)
- University Children's Hospital Tübingen (DE)
- University Hospital Leipzig (DE)
- Technical University of Munich (DE)
Publication Details
- Journal
- CardioVascular and Interventional Radiology
- Published
- 2025-12-07
- DOI
- https://doi.org/10.1007/s00270-025-04287-1
- Citations
- 1
- Primary Topic
- Topic Modeling
- Type
- article
- Field-Weighted Citation Impact
- 2.48