Skip to content
DraconiticResearch
Search Sources API
197,578 studies31 sourcesUpdated 52 min ago
Peer-reviewedCC BYArtificial Intelligence Review

Retrieval-augmented generation in biomedicine: a survey of technologies, datasets, and clinical applications

  • Jiawei He1,2,3
  • Boya Zhang2,3
  • Hossein Rouhizadeh2,3
  • Yingjian Chen4,5
  • Rui Yang6,7
  • Jin suo Lu8,9
  • Xudong Chen1,10
  • Nan Liu6,7
  • Douglas Teodoro2,3
10 affiliations
  1. 1Hunan City University, Yiyang, Hunan, China
  2. 2University of Geneva, Geneva, Switzerland
  3. 3University of Geneva
  4. 4Henan University, Kaifeng, Henan, China
  5. 5Henan University
  6. 6National University of Singapore, Singapore, Singapore
  7. 7National University of Singapore
  8. 8LvZhiDao Information Technology Co., Ltd, Changsha, Hunan, China
  9. 9LvZhiDao Information Technology Co., Ltd
  10. 10Hunan City University
Open at sourceDOICite

Abstract

Large language models (LLMs) in biomedicine face a fundamental conflict between static parameter knowledge and the dynamic nature of clinical evidence. Retrieval-Augmented Generation (RAG) addresses this by grounding generation in external data, yet it introduces new complexities in latency and architecture. This survey synthesizes the biomedical RAG landscape (2020–2025), classifying systems into naive, advanced, and modular paradigms. Beyond a technological taxonomy, we formalize the biomedical RAG trilemma, identifying the inherent trade-offs between reasoning depth, inference latency, and data privacy that constrain current clinical deployment. We analyze how recent agentic workflows enhance diagnostic reasoning but risk prohibitive latency, and how privacy constraints dictate the choice between powerful cloud-based models and local deployment. Finally, we outline the alignment gap in multimodal RAG and propose future directions for self-correcting, verifiable clinical agents.

Related

  1. Retrieval-augmented generation in medicine: A scoping review of technical implementations, clinical applications, and ethical considerations
    Rui Yang, Matthew Yu Heng Wong, Huitao Li +13
    Cell Reports MedicinePeer-reviewedCC BY-NC-ND18 citedEpidemiology & public healthOpenAlex
  2. Nanobodies as next‐generation targeting platforms: From discovery technologies to translational biomedicine
    Zhenrui Ye, Xianyang Li, Meixiao Zhan
    ViewPeer-reviewedCC BYFull textMedical imaging & radiologyDOAJ
  3. IPF-RAG: temporal-aware retrieval-augmented generation for idiopathic pulmonary fibrosis care.
    Xuejia Zheng, Xiang Rui, Teng Long +8
    Frontiers in digital healthPeer-reviewedCC BYFull textEpidemiology & public healthEurope PMCPMC
  4. Limits of Single-Pass Retrieval-Augmented Generation for AI-Powered Cancer Care Navigation: A Comparison of Retrieval Strategies
    E. Hasan, Y. Zhang, O. Cook +10
    medRxivPreprintCC BY-NC-NDFull textEpidemiology & public healthmedRxiv
  5. Integrating knowledge graphs and retrieval-augmented generation to identify clinical pharmacology gaps in infectious disease trial protocols
    Shreyas Waikar, Murali Ramanathan
    Frontiers in PharmacologyPeer-reviewedCC BY 4.0Full textEpidemiology & public healthCrossrefEurope PMCPMC

Cite

BibTeXRIS
BibTeX
@article{he2026retrievalaugmented,
  title = {{Retrieval-augmented generation in biomedicine: a survey of technologies, datasets, and clinical applications}},
  author = {Jiawei He and Boya Zhang and Hossein Rouhizadeh and Yingjian Chen and Rui Yang and Jin suo Lu and Xudong Chen and Nan Liu and Douglas Teodoro},
  journal = {Artificial Intelligence Review},
  year = {2026},
  month = oct,
  doi = {10.1007/s10462-026-11743-y},
  url = {https://doi.org/10.1007/s10462-026-11743-y},
  note = {CC BY},
}
RIS
TY  - JOUR
TI  - Retrieval-augmented generation in biomedicine: a survey of technologies, datasets, and clinical applications
AU  - Jiawei He
AU  - Boya Zhang
AU  - Hossein Rouhizadeh
AU  - Yingjian Chen
AU  - Rui Yang
AU  - Jin suo Lu
AU  - Xudong Chen
AU  - Nan Liu
AU  - Douglas Teodoro
PY  - 2026
DA  - 2026/10/23
JO  - Artificial Intelligence Review
DO  - 10.1007/s10462-026-11743-y
UR  - https://doi.org/10.1007/s10462-026-11743-y
AB  - Large language models (LLMs) in biomedicine face a fundamental conflict between static parameter knowledge and the dynamic nature of clinical evidence. Retrieval-Augmented Generation (RAG) addresses this by grounding generation in external data, yet it introduces new complexities in latency and architecture. This survey synthesizes the biomedical RAG landscape (2020–2025), classifying systems into naive, advanced, and modular paradigms. Beyond a technological taxonomy, we formalize the biomedical RAG trilemma, identifying the inherent trade-offs between reasoning depth, inference latency, and data privacy that constrain current clinical deployment. We analyze how recent agentic workflows enhance diagnostic reasoning but risk prohibitive latency, and how privacy constraints dictate the choice between powerful cloud-based models and local deployment. Finally, we outline the alignment gap in multimodal RAG and propose future directions for self-correcting, verifiable clinical agents.
SN  - 0269-2821
SN  - 1573-7462
ER  -