Comparison of proprietary and fine-tuned large language models for multi-label classification of billing codes from radiology reports
Question LLMs with high parameters possess medical knowledge, but how effective are they at predicting billing codes from radiology reports compared to smaller, fine-tuned models? Finidngs A fine-tuned ensemble model achieved competitive results and can outperform larger, proprietary LLMs. Clinical relevance Smaller, fine-tuned models offer an efficient alternative to proprietary LLMs in generating billing codes and can be integrated to assist clinical coding. This technology has the potential to transform clinical billing procedures, but its use should be overseen by qualified professional personnel.
Authors
- René Hosch (ORCID: https://orcid.org/0000-0003-1760-2342)
- Johannes Haubold (ORCID: https://orcid.org/0000-0003-4843-5911)
- Eva Maria Hartmann (ORCID: https://orcid.org/0009-0000-2600-7217)
- Cynthia Sabrina Schmidt (ORCID: https://orcid.org/0000-0003-1994-0687)
- Lale Umutlu (ORCID: https://orcid.org/0000-0001-5215-7171)
- Michael Forsting (ORCID: https://orcid.org/0000-0002-5584-9824)
- Katarzyna Borys (ORCID: https://orcid.org/0000-0001-6987-6041)
- Ahmad Idrissi-Yaghir (ORCID: https://orcid.org/0000-0003-1507-9690)
- Kamyar Arzideh (ORCID: https://orcid.org/0009-0005-6074-804X)
- Sina Warmer (ORCID: https://orcid.org/0009-0002-2262-2655)
- Bahadir Eryilmaz
- Felix Nensa
- Henning Schäfer
Institutions
- Ruhrlandklinik (DE)
- Essen University Hospital (DE)
Publication Details
- Journal
- European Radiology
- Published
- 2026-03-14
- DOI
- https://doi.org/10.1007/s00330-026-12445-3
- Primary Topic
- Topic Modeling
- Type
- article
- Field-Weighted Citation Impact
- 0.00