A multi-query, multimodal, receiver-augmented solution to extract contemporary cardiology guideline information using large language models
Our system demonstrated superior performance in question-answering tasks on a set of core cardiology questions as compared with contemporary publicly available chat models. The current study illustrates that LLMs can be tailored to provide documented and accountable guideline recommendations towards actual clinical needs while ensuring recommendations are derived from up-to-date, trustable, and traceable documents.
Authors
- Shelby Kutty (ORCID: https://orcid.org/0000-0001-9428-0979)
- Gerhard‐Paul Diller (ORCID: https://orcid.org/0000-0003-3050-5248)
- Rohan G. Reddy (ORCID: https://orcid.org/0009-0002-1993-9459)
- David A. Danford (ORCID: https://orcid.org/0000-0003-3344-3484)
- Pushpa Shivaram (ORCID: https://orcid.org/0000-0003-3661-8059)
- Robert Radke (ORCID: https://orcid.org/0000-0002-5055-2220)
Institutions
- Johns Hopkins University (US)
- Augusta University (US)
- Johns Hopkins Medicine (US)
- Harefield Hospital (GB)
- Augusta University Health (US)
- Lung Institute (US)
- University Hospital Münster (DE)
- BayCare Health System (US)
Publication Details
- Journal
- European Heart Journal - Digital Health
- Published
- 2025-09-20
- DOI
- https://doi.org/10.1093/ehjdh/ztaf111
- Primary Topic
- Biomedical Text Mining and Ontologies
- Type
- article
- Field-Weighted Citation Impact
- 0.00