Open-Source Large Language Models in Radiology: A Review and Tutorial for Practical Research and Clinical Deployment
dc.contributor.author | Savage, Cody H. | |
dc.contributor.author | Kanhere, Adway | |
dc.contributor.author | Parekh, Vishwa | |
dc.contributor.author | Langlotz, Curtis P. | |
dc.contributor.author | Joshi, Anupam | |
dc.contributor.author | Huang, Heng | |
dc.contributor.author | Doo, Florence X. | |
dc.date.accessioned | 2025-03-11T14:42:41Z | |
dc.date.available | 2025-03-11T14:42:41Z | |
dc.date.issued | 2025-01-28 | |
dc.description.abstract | Integrating large language models (LLMs) into health care holds substantial potential to enhance clinical workflows and care delivery. However, LLMs also pose serious risks if integration is not thoughtfully executed, with complex challenges spanning accuracy, accessibility, privacy, and regulation. Proprietary commercial LLMs (eg, GPT-4 [OpenAI], Claude 3 Sonnet and Claude 3 Opus [Anthropic], Gemini [Google]) have received much attention from researchers in the medical domain, including radiology. Interestingly, open-source LLMs (eg, Llama 3 and LLaVA-Med) have received comparatively little attention. Yet, open-source LLMs hold several key advantages over proprietary LLMs for medical institutions, hospitals, and individual researchers. The wider adoption of open-source LLMs has been slower, perhaps in part due to the lack of familiarity, accessible computational infrastructure, and community-built tools to streamline their local implementation and customize them for specific use cases. Thus, this article provides a tutorial for the implementation of open-source LLMs in radiology, including examples of commonly used tools for text generation and techniques for troubleshooting issues with prompt engineering, retrieval-augmented generation, and fine-tuning. Implementation-ready code for each tool is provided at https://github.com/UM2ii/Open-Source-LLM-Tools-for-Radiology. In addition, this article compares the benefits and drawbacks of open-source and proprietary LLMs, discusses the differentiating characteristics of popular open-source LLMs, and highlights recent advancements that may affect their adoption. | |
dc.description.sponsorship | F.X.D. is supported in part by an Association of Academic Radiology Clinical Effectiveness in Radiology Research Academic Fellowship Award and a grant from the Johns Hopkins Mid-Atlantic Center for Cardiometabolic Health Equity, which is supported by the National Institute on Minority Health and Health Disparities (P50MD017348). The content is solely theresponsibility of the authors and does not necessarily represent the official views of the Mid-Atlantic Center for Cardiometabolic Health Equity or the National Institutes of Health. C.P.L.is supported in part by the Medical Imaging and Data Resource Center, which is funded by the National Institute of Biomedical Imaging and Bioengineering (75N92020D00021). H.H.is partially supported by the National Institute on Aging (U01 AG068057), National Institute of Biomedical Imaging and Bioengineering (R01 EB034116), National Institute of GeneralMedical Sciences (R01 GM148743, R01 GM141076), and National Science Foundation (IIS 2347592, 2347604, 2348159, 2348169, DBI 2405416, CCF 2348306, CNS 2347617).H.H. and F.X.D. are supported in part by the Montgomery County, Maryland, and University of Maryland Strategic Partnership (MPowering the State), a formal collaboration between theUniversity of Maryland, College Park, and the University of Maryland, Baltimore | |
dc.description.uri | https://pubs.rsna.org/doi/full/10.1148/radiol.241073 | |
dc.format.extent | 18 pages | |
dc.genre | journal articles | |
dc.identifier | doi:10.13016/m2u1qz-gzon | |
dc.identifier.citation | Savage, Cody H., Adway Kanhere, Vishwa Parekh, Curtis P. Langlotz, Anupam Joshi, Heng Huang, and Florence X. Doo. "Open-Source Large Language Models in Radiology: A Review and Tutorial for Practical Research and Clinical Deployment". Radiology 314, no. 1 (January 2025): e241073. https://doi.org/10.1148/radiol.241073. | |
dc.identifier.uri | https://doi.org/10.1148/radiol.241073 | |
dc.identifier.uri | http://hdl.handle.net/11603/37762 | |
dc.language.iso | en_US | |
dc.publisher | RSNA | |
dc.relation.isAvailableAt | The University of Maryland, Baltimore County (UMBC) | |
dc.relation.ispartof | UMBC Center for Cybersecurity | |
dc.relation.ispartof | UMBC Faculty Collection | |
dc.relation.ispartof | UMBC Computer Science and Electrical Engineering Department | |
dc.rights | This item is likely protected under Title 17 of the U.S. Copyright Law. Unless on a Creative Commons license, for uses protected by Copyright Law, contact the copyright holder or the author. | |
dc.subject | Large Language Models | |
dc.subject | UMBC Cybersecurity Institute | |
dc.subject | Prompt Engineering | |
dc.subject | Retrieval-Augmented Generation | |
dc.subject | Radiology | |
dc.subject | Computational Infrastructure | |
dc.title | Open-Source Large Language Models in Radiology: A Review and Tutorial for Practical Research and Clinical Deployment | |
dc.type | Text | |
dcterms.creator | https://orcid.org/0000-0002-8641-3193 |