Assessing large language models for Lugano classification of malignant lymphoma in Japanese FDG-PET reports
GPT-4o outperformed other LLMs in assigning Lugano classification from Japanese FDG-PET free-text reports. This demonstrated the potential of advanced LLMs to interpret clinical texts. While the immediate clinical utility of automatically predicting a Lugano stage from an existing report may be limited, these results highlight the value of LLMs for understanding and standardizing free-text radiology data.
Authors
- Ryogo Minamimoto (ORCID: https://orcid.org/0000-0003-1026-255X)
- Keita Kato (ORCID: https://orcid.org/0000-0002-2901-2502)
- Shinji Naganawa
- Toshiaki Taoka (ORCID: https://orcid.org/0000-0001-9227-0240)
- Yumi Abe (ORCID: https://orcid.org/0000-0002-3881-0223)
- Rintaro Ito (ORCID: https://orcid.org/0000-0002-7971-043X)
- Katsuhiko Kato (ORCID: https://orcid.org/0000-0002-2083-9319)
- Hiroshi Ogawa (ORCID: https://orcid.org/0000-0001-7916-5622)
- Kosuke Nanataki
Institutions
- Nagoya University (JP)
Publication Details
- Journal
- EJNMMI Reports
- Published
- 2025-03-09
- DOI
- https://doi.org/10.1186/s41824-025-00246-8
- Citations
- 3
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 4.08
Funders
- Japan Society for the Promotion of Science