Streamlining systematic reviews with large language models using prompt engineering and retrieval augmented generation
The LLM-based system significantly enhances SR efficiency, compared to manual methods and Rayyan while maintaining low FNR.
Authors
- Maya Rahme (ORCID: https://orcid.org/0000-0002-9467-5062)
- Ryan Yammine (ORCID: https://orcid.org/0009-0005-9183-7395)
- Marlene Chakhtoura (ORCID: https://orcid.org/0000-0003-0012-986X)
- Ghada El‐Hajj Fuleihan (ORCID: https://orcid.org/0000-0002-2076-5858)
- Fouad Trad (ORCID: https://orcid.org/0000-0003-2241-8195)
- Jana Charafeddine
- Ali Chehab
Institutions
- American University of Beirut (LB)
Publication Details
- Journal
- BMC Medical Research Methodology
- Published
- 2025-05-10
- DOI
- https://doi.org/10.1186/s12874-025-02583-5
- Citations
- 20
- Primary Topic
- Meta-analysis and systematic reviews
- Type
- article
- Field-Weighted Citation Impact
- 27.97
Funders
- Office of Dietary Supplements
- National Institutes of Health
- Fogarty International Center