Assessing the ability of large language models to simplify lumbar spine imaging reports into patient-facing text: a pilot study of GPT-4
GPT-4 effectively simplifies the reading level of lumbar spine MRI reports. The model tends to omit key information in its translations, which can be mitigated with enhanced prompting. Further validation in the domain of spine radiology needs to be performed to facilitate clinical integration.
Authors
- Rushmin Khazanchi
- Daniel Herrera (ORCID: https://orcid.org/0009-0009-3487-3073)
- Srikanth N. Divi (ORCID: https://orcid.org/0000-0002-5776-2044)
- Austin R Chen
- Mykhaylo Krushelnytskyy (ORCID: https://orcid.org/0000-0002-2561-6907)
- Hanna Kemeny
- Parth Desai (ORCID: https://orcid.org/0000-0003-1231-1204)
- Jacob Staub (ORCID: https://orcid.org/0009-0007-2576-1446)
- Wellington K. Hsu (ORCID: https://orcid.org/0000-0001-9086-6868)
- Alpesh A. Patel (ORCID: https://orcid.org/0000-0003-2193-4294)
- Matthew Follett (ORCID: https://orcid.org/0000-0002-5352-7119)
Institutions
- Northwestern University (US)
- Neurological Surgery (US)
Publication Details
- Journal
- Skeletal Radiology
- Published
- 2025-09-09
- DOI
- https://doi.org/10.1007/s00256-025-05027-9
- Primary Topic
- Artificial Intelligence in Healthcare and Education
- Type
- article
- Field-Weighted Citation Impact
- 0.00