Assessing Large Language Models for Oncology Data Inference From Radiology Reports
LLMs, especially GPT-4, are proficient in deriving oncologic insights from radiology reports. Their performance is enhanced by effective summarization strategies, demonstrating their potential in clinical support and health care analytics. This study also underscores the possibility of zero-shot open model utility in environments where proprietary models are restricted. Finally, by providing a set of annotated radiology reports, this paper presents a valuable data set for further LLM research in oncology.
Authors
- Madhumita Sushil (ORCID: https://orcid.org/0000-0001-7884-0526)
- Julian C. Hong (ORCID: https://orcid.org/0000-0001-5172-6889)
- Corynn Kasap (ORCID: https://orcid.org/0000-0002-3667-8097)
- Travis Zack (ORCID: https://orcid.org/0000-0002-1620-6455)
- Brenda Y. Miao (ORCID: https://orcid.org/0000-0002-3393-9837)
- Atul J. Butte (ORCID: https://orcid.org/0000-0002-7433-2740)
- L. Chen (ORCID: https://orcid.org/0000-0002-5162-3756)
- Eric A. Collisson (ORCID: https://orcid.org/0000-0002-2829-355X)
- Arda Demirci (ORCID: https://orcid.org/0000-0002-0245-0334)
Institutions
- University of California, San Francisco (US)
- UCSF Helen Diller Family Comprehensive Cancer Center (US)
- University of California, Berkeley (US)
Publication Details
- Journal
- JCO Clinical Cancer Informatics
- Published
- 2024-12-01
- DOI
- https://doi.org/10.1200/cci.24.00126
- Citations
- 13
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 5.26