An open-source fine-tuned large language model for radiological impression generation: a multi-reader performance study
An open-source fine-tuned LLM can generate impressions to a satisfactory level of clinical accuracy, grammatical accuracy, and stylistic quality. Our reader performance study demonstrates the potential of large language models in drafting radiology report impressions that can aid in streamlining radiologists' workflows.
Authors
- Jamie Schroeder (ORCID: https://orcid.org/0000-0002-3710-911X)
- Maya Vella (ORCID: https://orcid.org/0000-0002-4816-5563)
- Jae Ho Sohn (ORCID: https://orcid.org/0000-0002-6733-7551)
- Adrian Serapio
- Gunvant Chaudhari (ORCID: https://orcid.org/0000-0003-2872-3828)
- Cody Savage (ORCID: https://orcid.org/0000-0001-6854-1397)
- Shravan Sridhar (ORCID: https://orcid.org/0000-0002-5934-2166)
- Jonathan Liu (ORCID: https://orcid.org/0000-0001-5650-3086)
- Adam Yala
- Yoo Jin Lee (ORCID: https://orcid.org/0009-0008-2348-2520)
Institutions
- University of Maryland, Baltimore (US)
- University of Maryland Medical Center (US)
- University of California, San Francisco (US)
- Georgetown University (US)
- University of Washington (US)
- MedStar Georgetown University Hospital (US)
- University of California, Berkeley (US)
Publication Details
- Journal
- BMC Medical Imaging
- Published
- 2024-09-27
- DOI
- https://doi.org/10.1186/s12880-024-01435-w
- Citations
- 22
- Primary Topic
- Radiology practices and education
- Type
- article
- Field-Weighted Citation Impact
- 8.89