Retrieval augmented generation for large language models in healthcare: A systematic review
Open accessThe DOI matches its Crossref record; 161 of the 161 references it cites were found in a registry or an index.
Abstract
Large Language Models (LLMs) have demonstrated promising capabilities to solve complex tasks in critical sectors such as healthcare. However, LLMs are limited by their training data which is often outdated, the tendency to generate inaccurate (“hallucinated”) content and a lack of transparency in the content they generate. To address these limitations, retrieval augmented generation (RAG) grounds the responses of LLMs by exposing them to external knowledge sources. However, in the healthcare domain there is currently a lack of systematic understanding of which datasets, RAG methodologies and evaluation frameworks are available. This review aims to bridge this gap by assessing RAG-based approaches employed by LLMs in healthcare, focusing on the different steps of retrieval, augmentation and generation. Additionally, we identify the limitations, strengths and gaps in the existing literature. Our synthesis shows that 78.9% of studies used English datasets and 21.1% of the datasets are in Chinese. We find that a range of techniques are employed RAG-based LLMs in healthcare, including Naive RAG, Advanced RAG, and Modular RAG. Surprisingly, proprietary models such as GPT-3.5/4 are the most used for RAG applications in healthcare. We find that there is a lack of standardised evaluation frameworks for RAG-based applications. In addition, the majority of the studies do not assess or address ethical considerations related to RAG in healthcare. It is important to account for ethical challenges that are inherent when AI systems are implemented in the clinical setting. Lastly, we highlight the need for further research and development to ensure responsible and effective adoption of RAG in the medical domain.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Figures
![Fig 1 PRISMA [25] workflow applied to the search, identification and selection of the studies that were included in the systematic review.](https://figures.cochat.ai/paper-figures/558117d130e94385a047aeb76b2922c7/retrieval-augmented-generation-for-large-language-models-in-healthcare-a-figure-1.jpg)



Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source says what the claim attributes to it.
In the citing paper
“Some general-purpose LLMs, including GPT-4-based models, can be customized through methods such as prompt engineering and in-context learning (e.g., PharmacyGPT), 10 and augmented with domain-specific external knowledge sources, enabling responses grounded in relevant databases or documents.114”
From Large Language Models in Pharmacist-Led Direct Patient Care: A Rapid Review of Published Evaluations (2026) · cites this paper as [114]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018) · arXiv · doi:10.48550/arxiv.1810.04805 · Cited by 73,173 · cited 1×
“Models such as BioBERT [8] and ClinicalBERT [9] have been proposed, leveraging the power of Bidirectional Encoder Representations from Transformers (BERT) [10].”
XGBoost: A Scalable Tree Boosting System
Chen, T., & Guestrin, C. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2939785 · Cited by 54,569 · cited 1×
“A recent study by Jin et al. [69] proposed a framework that integrates advanced techniques such as large-scale feature extraction combined with RAG, accurate scoring of features from medical knowledge using LlamaIndex, and XGBoost [131] algorithm for prediction.”
Liberati, A., Altman, D. G., Tetzlaff, J., Mulrow, C., Gøtzsche, P. C., Ioannidis, J. P. A., Clarke, M., Devereaux, P. J., Kleijnen, J., & Moher, D. (2009) · PLoS Medicine · doi:10.1371/journal.pmed.1000100 · Cited by 28,164 · cited 1×
“All studies included met the inclusion criteria shown in Table 1.”
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., Zhou, Y., Li, W., & Liu, P. J. (2019) · arXiv · doi:10.48550/arxiv.1910.10683 · Cited by 10,929
BioBERT: a pre-trained biomedical language representation model for biomedical text mining
Lee, J., Yoon, W., Kim, S., Kim, D., Kim, S., So, C. H., & Kang, J. (2020) · Bioinformatics · doi:10.1093/bioinformatics/btz682 · Cited by 7,218 · cited 1×
“Models such as BioBERT [8] and ClinicalBERT [9] have been proposed, leveraging the power of Bidirectional Encoder Representations from Transformers (BERT) [10].”
Survey of Hallucination in Natural Language Generation
Ji, Z., Lee, N., Frieske, R., Yu, T., Su, D., Xu, Y., Ishii, E., Bang, Y. J., Madotto, A., & Fung, P. (2023) · ACM Computing Surveys · doi:10.1145/3571730 · Cited by 5,025 · cited 1×
“This issue is often referred to as "model hallucinations" [6] and methodologies for its mitigation are still an active area of research [7].”
Bag of Tricks for Efficient Text Classification
Joulin, A., Grave, E., Bojanowski, P., & Mikolov, T. (2017) · Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 2, Short Papers · doi:10.18653/v1/e17-2068 · Cited by 4,545
Bag of Tricks for Efficient Text Classification
Joulin, A., Grave, E., Bojanowski, P., & Mikolov, T. (2016) · arXiv · doi:10.48550/arxiv.1607.01759 · Cited by 4,542
Large language models encode clinical knowledge
Singhal, K., Azizi, S., Tu, T., Mahdavi, S. S., Wei, J., Chung, H. W., Scales, N., Tanwani, A., Cole-Lewis, H., Pfohl, S., et al. (2023) · Nature · doi:10.1038/s41586-023-06291-2 · Cited by 4,083 · cited 1×
“To address this contextual need in medicine, Med-Palm was introduced demonstrating good performance in retrieving clinical knowledge and excelling in decision-making on several clinical tasks [11].”
Research works printed but not linked: 4
- Abacha AB, Agichtein E, Pinter Y, Demner-Fushman D. Overview of the medical question answering task at TREC 2017 LiveQA. In: Text Retrieval Conference; 2017. Available from: https://api.semanticscholar.org/CorpusID:3902472
- Wang J, Yang Z, Yao Z, Yu H. JMLR: joint medical LLM and retrieval training for enhancing reasoning and professional question answering capability. 2024.
- H Yu, 2023, Zero-shot ECG diagnosis with large language models and retrieval-augmented generation., 650
- DD Team, 2024, DeepEval: the open-source LLM evaluation framework
Cited by
Retrieval-Augmented Generation (RAG) Chatbots for Education: A Survey of Applications
Swacha, J., & Gracel, M. (2025) · Applied Sciences · doi:10.3390/app15084234 · Cited by 244
Hindi, M., Mohammed, L., Maaz, O., & Alwarafy, A. (2025) · IEEE Access · doi:10.1109/access.2025.3550145 · Cited by 140
AI Agents in Clinical Medicine: A Systematic ReviewPreprint
Gorenshtein, A., Omar, M., Glicksberg, B. S., Nadkarni, G. N., & Klang, E. (2025) · medRxiv · doi:10.1101/2025.08.22.25334232 · Cited by 94
“14,15 20, 21, 22 Unlike standard”
The rise of small language models in healthcare: A comprehensive survey
Garg, M., Raza, S., Rayana, S., Liu, X., & Sohn, S. (2026) · Computer Science Review · doi:10.1016/j.cosrev.2026.100999 · Cited by 81
A Current Review of Generative AI in Medicine: Core Concepts, Applications, and Current Limitations
Rouzrokh, P., Khosravi, B., Faghani, S., Moassefi, M., Shariatnia, M. M., Rouzrokh, P., & Erickson, B. J. (2025) · Current Reviews in Musculoskeletal Medicine · doi:10.1007/s12178-025-09961-y · Cited by 66
“RAG models provide significant advancements in over - coming the limitations of traditional large language models (LLMs), where accuracy and timeliness are critical [ 83 ].”
General-purpose large language models outperform specialized clinical AI tools on medical benchmarks
Vishwanath, K., Alyakin, A., Ghosh, M., Hage, A., Neifert, S. N., Orillac, C., Mandelberg, N. J., Khan, H. A., Lee, J. V., Yao, J. J., et al. (2026) · Nature Medicine · doi:10.1038/s41591-026-04431-5 · Cited by 49
Large language models in clinical trials: applications, technical advances, and future directions
Lin, A.-Q., Wang, Z., Jiang, A.-M., Chen, L., Qi, C., Zhu, L., Mou, W., Gan, W.-Y., Zeng, D.-Q., Xiao, M.-J., et al. (2025) · BMC Medicine · doi:10.1186/s12916-025-04348-9 · Cited by 35
“The current strategy for updating the knowledge base mainly involves leveraging external retrieval information to override outdated content in the model’s output, such as data augmentation and internet augmentation [117].”
Wang, S., Yang, H., & Liu, W. (2025) · Scientific Reports · doi:10.1038/s41598-025-21222-z · Cited by 31
“In addition, Amugongo et al. (2024) 8 conducted a systematic review of RAG methods in healthcare, highlighting three major challenges in KG-RAG models: (1) weak integration between structured and unstructured knowledge, (2) limited accuracy in selecting multi-hop reasoning paths, and (3) lack of…”
Livingston, L., Featherstone-Uwague, A., Barry, A., Barretto, K., Morey, T., Herrmannová, D., & Avula, V. (2025) · JAMIA Open · doi:10.1093/jamiaopen/ooaf054 · Cited by 29
AlzheimerRAG: Multimodal Retrieval-Augmented Generation for Clinical Use Cases
Lahiri, A. K., & Hu, Q. (2025) · Machine Learning and Knowledge Extraction · doi:10.3390/make7030089 · Cited by 31
Younesi, A., Oustad, E., Ansari, M., Fahringer, T., & Buyya, R. (2025) · Internet of Things · doi:10.1016/j.iot.2025.101815 · Cited by 34
kizi, M. K. Z., & Suh, Y. (2025) · Electronics · doi:10.3390/electronics14153095 · Cited by 26
POC Sensor Systems and Artificial Intelligence—Where We Are Now and Where We Are Going?
Prashanthi, K., Kovur, K. M., Rayat, D. Y., & Wishart, D. S. (2025) · Biosensors · doi:10.3390/bios15090589 · Cited by 25
“The integration of LLMs with RAG systems represents a promising advancement in the development of ML-augmented DCTs [169–172].”
Bunnell, D. J., Bondy, M. J., Fromtling, L. M., Ludeman, E. M., & Gourab, K. (2025) · medRxiv · doi:10.1101/2025.04.01.25325033 · Cited by 22
Wołk, K. (2025) · Electronics · doi:10.3390/electronics14214227 · Cited by 27
“The clinical retrieval benchmark datasets consist of TREC Precision Medicine [14] and MedMCQA [16] and the newly developed ClinRAG-100K, which offers 100,000 patient–question–answer triples with PubMed PMIDs for detailed evaluation [24].”
LatteReview: A Multi-Agent Framework for Systematic Review Automation Using Large Language Models
Rouzrokh, P., Khosravi, B., Rouzrokh, P., & Shariatnia, M. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2501.05468 · Cited by 19
Yang, R., Wong, M. Y. H., Li, H., Li, X., Zhu, W., Liao, J., Yu, K., Liew, J. C. K., Xuan, W., Chen, Y., et al. (2026) · Cell Reports Medicine · doi:10.1016/j.xcrm.2026.102927 · Cited by 28
EvoMDT: a self-evolving multi-agent system for structured clinical decision-making in multi-cancer
Liu, Q., Hu, Z., Huang, T., Niu, Y., Zhang, X., Ma, S., Lin, C., Goh, K. H., Kwon, H. E., Gao, F., et al. (2026) · npj Digital Medicine · doi:10.1038/s41746-025-02304-8 · Cited by 31
Voultsiou, E., & Moussiades, L. (2026) · Electronics · doi:10.3390/electronics15030524 · Cited by 27
From Traditional Machine Learning Models to Multimodal Large Models: A Review of Aquaculture
Liu, S., Du, Z., Wang, G., Zhang, P., Xu, W., Yu, J., & Li, D. (2025) · Reviews in Aquaculture · doi:10.1111/raq.70111 · Cited by 19
Stuhlmiller, T. J., Rabe, A., Rapp, J., Awawda, A., Kouser, H., Lui, K., Salamon, H., Chuyka, D., Mahoney, W., Furgason, J. M., et al. (2025) · medRxiv · doi:10.1101/2025.02.25.25322898 · Cited by 13
Guo, Z., Lai, A. G., Ive, J., Petcu, A. P., Wang, Y., Qi, L., Thygesen, J. H., & Li, K. (2026) · PLOS Digital Health · doi:10.1371/journal.pdig.0001446 · Cited by 16
“…clarifications and feedback in authoritative psychological and helpline sources, this design aims to improve the reliability and interpretability of chatbot responses and to mitigate potential hallucinations, defined as the generation of plausible but factually unsup-ported outputs [29,30].”
Prabha, S., Gomez-Cabello, C. A., Haider, S. A., Genovese, A., Trabilsy, M., Wood, N. G., Bagaria, S. P., Tao, C., & Forte, A. J. (2025) · Bioengineering · doi:10.3390/bioengineering12080895 · Cited by 14
“This is not different when it comes to RAG; performance is sensitive to retrieval design choices, and evaluation is not yet standardized [43].”
A Systematic Literature Review of Artificial Intelligence in Prehospital Emergency Care
Elfahim, O., Edjinedja, K. L., Cossus, J., Youssfi, M., Barakat, O., & Desmettre, T. (2025) · Big Data and Cognitive Computing · doi:10.3390/bdcc9090219 · Cited by 14
“Studies suggest numerous strategies to reduce hallucinations and improve factual accuracy, strategies like retrieval-augmented generation (RAG) [87], agentic AI [88], and prompt engineering [89].”
RAGMed: A RAG-Based Medical AI Assistant for Improving Healthcare Delivery
Patil, R., Abbidi, M., & Fannon, S. (2025) · AI · doi:10.3390/ai6100240 · Cited by 14
Effective prompt design for large language models in clinical practice
Callens, S. F. J. (2026) · Acta Clinica Belgica · doi:10.1080/17843286.2026.2613903 · Cited by 14
“Retrieval-augmented generation (RAG) systems can ground responses in verified sources and provide citations, reducing hallucinations by over 40% in some frameworks, though standardized evaluation remains lacking [33,52].”
Mondal, H., Dash, I., Mondal, S., Varikasuvu, S. R., Gayen, R. K., Sharma, S., & Biri, S. K. (2025) · Expert Review of Clinical Pharmacology · doi:10.1080/17512433.2025.2568090 · Cited by 14
“These models have demonstrated capabilities in medical question answering [5], clinical decision support [6], and biomedical knowledge retrieval [7].”
Imran, M., & Lee, Y. (2026) · IEEE Access · doi:10.1109/access.2026.3656368 · Cited by 15
Information-Constrained Retrieval for Scientific Literature via Large Language Model Agents
Zheng, J., Chen, Y., Zhou, Z., Peng, C., Deng, H., & Yin, S. (2025) · International Conference on Big Data, Artificial Intelligence and Internet of Things Engineering (ICBAIE) · doi:10.1109/icbaie66852.2025.11326646 · Cited by 12
“By introducing constraints into the retrieval process, the model can ensure high relevance while also improving the diversity and interpretability of results [23-25].”
Orchestrated multi agents sustain accuracy under clinical-scale workloads compared to a single agent
Klang, E., Omar, M., Raut, G., Agbareia, R., Timsina, P., Freeman, R., Gavin, N., Stump, L., Charney, A. W., Glicksberg, B. S., et al. (2026) · npj Health Systems · doi:10.1038/s44401-026-00077-0 · Cited by 5
“…a fi xed overhead (orchestrator + worker calls and repeated prompts), which can increase total tokens and latency at smaller batch sizes for some checkpoints; however, as batch size grows, task partitioning prevents context-driven degradation and yields more predictable scaling under load.”
Chen, H.-K., & Wang, H.-H. (2025) · Biomimetics · doi:10.3390/biomimetics10090626 · Cited by 8
“These findings parallel observations from healthcare RAG reviews [22–24], which emphasize that rigorous evaluation requires moving beyond retrieval accuracy to outcome-level measures that reflect domain-specific utility.”
InfluentialHamamoto, R., Koyama, T., Takahashi, S., Yasuda, T., Kobayashi, K., Akagi, Y., Kouno, N., Sudo, K., Hirata, M., Sunami, K., et al. (2026) · Journal of Hematology & Oncology · doi:10.1186/s13045-026-01781-y · Cited by 8
“…documents using LLMs [138, 171], grounding in trial registries and guidelines through RAG to verify information against the latest information [35, 169], confirming effectiveness through EHR standardization and multicenter prospective evaluation [172, 173], and establishing audit-able workflows…”
Application of knowledge graphs in rare disease research
Fei, Y., Ding, H., Tong, S., He, Y., & Cai, W. (2026) · Frontiers in Public Health · doi:10.3389/fpubh.2026.1757612 · Cited by 8
Olszewski, R., Brzeziński, J., Watros, K., & Rysz, J. (2026) · Applied Sciences · doi:10.3390/app16031423 · Cited by 9
Bai, B., Meng, X., & Zhao, C. (2025) · Applied Sciences · doi:10.3390/app151910850 · Cited by 7
Hajar, M. R., Utami, E., & Muhammad, A. H. (2025) · Journal of Applied Computer Science and Technology · doi:10.52158/jacost.v6i2.1170 · Cited by 7
Dehdab, R., Afat, S., Mankertz, F., Brendel, J. M., Maalouf, N., Werner, S., Brendlin, A., Herrmann, J., Nikolaou, K., Kloker, L. D., et al. (2026) · Journal of Cancer Research and Clinical Oncology · doi:10.1007/s00432-026-06432-w · Cited by 5
Kadi, H., Abdellatif, A., Njamen, D. I. K., & Pereme, F. (2025) · Frontiers in Medicine · doi:10.3389/fmed.2025.1598979 · Cited by 5
“Amugongo et al. (26) reviewed RAG in healthcare and emphasized the use of dense embeddings to enhance retrieval accuracy and reduce hallucinations.”
Retrieval-Augmented Generation Assistant for Anatomical Pathology Laboratories
Pires, D., Perezhohin, Y., & Castelli, M. (2025) · Emerging Science Journal · doi:10.28991/esj-2025-09-06-013 · Cited by 6
“Studies have shown RAG’s effectiveness in clinical documentation assistance, medical question answering, and diagnostic support [10, 11].”
InfluentialCollaço, B. G., Srinivasagam, P., Gomez-Cabello, C. A., Haider, S. A., Genovese, A., Wood, N. G., Bagaria, S. P., Lifson, M. A., & Forte, A. J. (2026) · Bioengineering · doi:10.3390/bioengineering13020225 · Cited by 8
“…as a safer alternative to parameter updates, Amugongo et al. (2025) cautions that data leakage remains possible through both the foundation model (pretraining/FT datasets) and the retrieval dataset, which in clinical settings may include diagnoses, medications, and private patient information [10].”
InfluentialMikkelsen, Y. (2026) · JMIR Medical Informatics · doi:10.2196/94241 · Cited by 7
“The 2025 systematic review of RAG in health care [10] identified “limited evaluation frameworks” as a persistent challenge; the factorial approach directly addresses this.”
Yu, Y. Q., Li, L., & Li, Y. (2025) · medRxiv · doi:10.1101/2025.10.17.25338266 · Cited by 6
Blašković, L., Tanković, N., Lorencin, I., & Šegota, S. B. (2025) · Big Data and Cognitive Computing · doi:10.3390/bdcc9100256 · Cited by 8
“Regarding healthcare, Amugongo et al. [58] reported that RAG has been primarily deployed in clinical decision support, medical QA, and the augmentation of EHRs.”
Mastroleo, F., Borras-Osorio, M., Patel, S. P., Peterson, S., Wilson, R., Zhou, M., Shiraishi, S., Foong, A. Y. K., Routman, D. M., & Waddle, M. R. (2025) · Radiotherapy and Oncology · doi:10.1016/j.radonc.2025.111348 · Cited by 4
Deng, L., Chen, L., Liu, M., Wang, X., Qi, Y., Shao, C., & Jiang, T. (2025) · Information Processing & Management · doi:10.1016/j.ipm.2025.104435 · Cited by 4
Developing an AI Assistant for Knowledge Management and Workforce Training in State DOTs
Amaram, D., Gao, L., Gudla, G. R., & Katale, T. S. (2026) · Electronics · doi:10.3390/electronics15010217 · Cited by 6
“Researchers are also examining advanced architectures, such as multi-agent or agentic RAG systems [21].”
Large language models for neurology: a mini review
Wunsch, D. C., & Hier, D. B. (2026) · Frontiers in Digital Health · doi:10.3389/fdgth.2025.1732759 · Cited by 4
“Retrieval-augmented generation (RAG) architectures provide a more scalable solution, coupling dynamic knowledge sources with stable reasoning engines [54, 55].”
VerifAI: A Verifiable Open-Source Search Engine for Biomedical Question Answering
Košprdić, M., Ljajić, A., Bašaragin, B., Medvecki, D., Cassano, L., & Milošević, N. (2026) · IEEE Access · doi:10.1109/access.2026.3676985 · Cited by 4
“A systematic review of retrieval-augmented generation for large language models in healthcare further provides evidence-based guidelines for implementing RAG systems in clinical settings [29].”
Artificial Intelligence in Head and Neck Surgical Oncology: A State-of-the-Art Review
Chen, S., Feucht, M., Bhatt, A. M., & Farlow, J. L. (2026) · Journal of Clinical Medicine · doi:10.3390/jcm15072767 · Cited by 5
“Systematic reviews show RAG often improves factuality but requires standardized evaluation and careful choice of retrieval methods and reference sources [19,20].”
Tezcan, H., Tunçez, A., Gürses, K. M., Özen, Y., & Yalçın, M. U. (2026) · BMC Medical Education · doi:10.1186/s12909-026-09141-7 · Cited by 5
Enhancing Medical Question Answering with LLMs via a Hybrid Retrieval-Augmented Generation Framework
Aljohani, B., & Alsanoosy, T. (2026) · Information · doi:10.3390/info17020133 · Cited by 3
Wang, J.-F., Chang, C.-C., Chiang, T.-M., Yeh, T., Cheng, E., Lee, Y., & Chen, H. (2025) · Health Informatics Journal · doi:10.1177/14604582251381233 · Cited by 4
“3,15,16 Retrieval-augmented generation (RAG) The core concept of RAG 17 is to combine external retrieval mechanisms with generative models to provide more accurate content.”
Large language model enhanced framework for systematic reviews and meta-analyses
Shen, J., Luo, Z.-Y., Jia, D., Wang, S.-R., Sun, F., & Wu, J.-G. (2025) · BMJ digital health & AI · doi:10.1136/bmjdhai-2025-000017 · Cited by 6
“37 We incorporate RAG to enhance factual accuracy and domain-specific reasoning by dynamically retrieving external documents (eg, abstracts, clinical trial reports or guidelines) and feeding them into the LLM’s prompt context.”
Modi, M., Krull, J. E., Gauntner, T. D., Li, M., Johnson, D., Wang, X., Cheng, H., Zhao, J., Ma, Q., Zhang, P., et al. (2026) · npj Digital Medicine · doi:10.64898/2026.01.26.26344845 · Cited by 4
Large Language Models and Otolaryngology: A Review
Bao, J. W., Jawad, M., Pavelchek, C., Powell, W. J. B., Oh, I. Y., Gupta, A., & Shew, M. A. (2026) · JAMA Otolaryngology–Head & Neck Surgery · doi:10.1001/jamaoto.2025.5335 · Cited by 6
Kaba, E., Çubukçu, Y., Uzunibrahimoğlu, B., Yılmaz, Y. E., Çınar, M., Tabakoğlu, S., Bal, E., Beydüz, Y. S., Solak, M., Oğuz, E. A., et al. (2026) · Die Radiologie · doi:10.1007/s00117-026-01580-z · Cited by 3
Retrieval-enhanced drafting of ClinicalTrials.gov data elements from clinical protocols
Baluguri, R. S., & Anderson, N. (2026) · Journal of Clinical and Translational Science · doi:10.1017/cts.2026.10735 · Cited by 3
Li, M., Chen, D., Xiao, Q., He, Z., Zhang, Y., Zhong, J., Luo, Y., & Ma, H. (2026) · Journal of Medical Internet Research · doi:10.2196/89850 · Cited by 3
“Third, as highlighted in recent health care RAG literature, robust evaluation frameworks are essential to mitigate issues related to bias, trust, and privacy [58].”
Retrieval-Augmented Generation in Oncology: Promises, Pitfalls, and Early Applications
Thaker, N. G., Liu, W., Waddle, M., Showalter, T. N., Mastroleo, F., Luh, J. Y., Levitt, C., Ning, M., Loaiza‐Bonilla, A., & Hong, J. C. (2026) · AI in Precision Oncology · doi:10.1177/2993091x261446348 · Cited by 3
“42 Additionally, the cost of RAG work fl ows may be higher than zero-shot work fl ows, due to the increased cost of vector embeddings (and associated data retrieval and indexing) and tokenization.”
Lim, Y. X., Ming, L. C., Ng, N. C., Leow, S., & Ingle, R. G. (2026) · Discover Computing · doi:10.1007/s10791-026-10205-x · Cited by 3
“Modern approaches often use advanced architectures like Bidirectional Encoder Representations from Transformers (BERT) and its clinical variants (e.g., BioBERT, UmlsBERT), Long Short-Term Memory (LSTM) networks, and Retrieval-Augmented Generation (RAG) systems [22–25].”
Mingole, B., Majumdar, A., Choudhury, F. A., Kraschnewski, J. L., Sundar, S. S., & Yadav, A. (2026) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3805689.3812315 · Cited by 2
Pambudi, S., & Menolascina, F. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2510.04969 · Cited by 3
“…well-adapted retrievers and standalone LLMs face limitations in tackling complex, real-world clinical tasks: single-agent models may misinterpret ambiguous inputs, hallucinate un-supported conclusions, or fail to integrate multiple sources effectively (Ozmen & Mathur, 2025; Amugongo et al., 2025).”
Akindele, O. K., Mishra, B. K., & Wertheim, K. Y. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2509.19209 · Cited by 3
Pocket RAG: On-Device RAG for First Aid Guidance in Offline Mobile EnvironmentPreprint
Kang, D. H., Lee, H., Cha, H., Choi, M., & Lim, S. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2602.13229 · Cited by 5
“A systematic review by Amugongo et al. [19] confirms that most healthcare RAG studies depend on cloud resources, making them unsuitable for disconnected environments.”
Emssaad, I., Ben-Bouazza, F.-E., Tafala, I., Mezali, M. C. E., & Jioudi, B. (2025) · Computer Methods and Programs in Biomedicine Update · doi:10.1016/j.cmpbup.2025.100221 · Cited by 2
Noguera, A., Mogollón-Benavides, A. L., Niño-Mojica, M. D., Rúa, S., Sanín-Villa, D., & Tejada, J. C. (2025) · The Scientist · doi:10.3390/sci7040148 · Cited by 2
“A systematic review published in 2025 further highlights the growing adoption of RAG in healthcare, while identifying persistent gaps in ethics and evaluation metrics within these systems [3].”
InfluentialJia, Q., Liu, S., Chen, M., Li, T., & Yang, J. (2025) · Tomography · doi:10.3390/tomography11100115 · Cited by 2
“The core idea of RAG is to combine the reasoning and generation capabilities of large language models with the accuracy of an external trustworthy knowledge base [39].”
Sesen, M. B., Yeung, J. A., & Asgari, E. (2026) · BMJ Digital Health & AI · doi:10.1101/2025.11.25.25341010 · Cited by 2
“These findings are consistent with recent research demonstrating that custom LLM models grounded in evidence can significantly improve clinical correctness for evidence-based medicine 31 .”
Enhancing Epidemiological Insights with RAG for SIREVA-SUS Reports
Freitas, C., Rabonato, R. T., & Berton, L. (2025) · doi:10.5753/eniac.2025.11796 · Cited by 2
“A review of the use of RAG in healthcare was conducted by [Amugongo et al. 2025], addressing the limitations of LLMs, such as outdated training data, hallucinated content, and lack of transparency.”
Large language models in drug delivery: A review of the current landscape and future perspectives
Sakkal, M., & Hajal, A. A. (2025) · Journal of Pharmaceutical Sciences · doi:10.1016/j.xphs.2025.104147 · Cited by 2
Beyond Autonomy: A Plastic Surgeon's Responsibility in the Face of AI-Driven Misinformation
Rajaram, P., Lane, M. E., Andalibi, N., Haimson, O. L., Hooper, R. C., & Prescher, H. (2025) · Annals of Plastic Surgery · doi:10.1097/sap.0000000000004608 · Cited by 3
Putra, P. K., Nurfadhilah, E., Pebiana, S., & Prakasa, E. (2025) · International Conference on Computer, Control, Informatics and its Applications · doi:10.1109/ic3ina68387.2025.11325125 · Cited by 2
Cimiano, P., Brazda, N., Härtung, M., Starke-Knäusel, C., Schmidt, A. L., Vuono, M. C. D., Tyagi, A. S., Gottowik, J., Rodriguez‐Esteban, R., Collins, B., et al. (2026) · Frontiers in Medicine · doi:10.3389/fmed.2025.1703923 · Cited by 2
“…a qualitative summary of the impact of depression on the daily activities of patients by using a retrieval-augmented generation (RAG) approach (94) and having an LLM automatically summarize the relevant documents describing the impact of depression, following the methodology proposed by Nair…”
Lin, S., shao, S., Liu, X., & Su, H. (2026) · Information · doi:10.3390/info17020138 · Cited by 2
Yadav, T., Tekale, A., Chong, J., & Masum, M. (2026) · Algorithms · doi:10.3390/a19030215 · Cited by 2
Language Models and Food–Health Evidence: Challenges, Opportunities, and Implications
Jackson, D., Gousiopoulos, A., & Soldatos, T. G. (2026) · BioMedInformatics · doi:10.3390/biomedinformatics6020013 · Cited by 3
“A systematic review of 70 healthcare RAG studies found that hallucination rates decreased in all cases where this was directly measured, though most evaluations were limited to English-and Chinese-language data and proprietary models [93].”
Large language models for primary care ophthalmic education: a systematic review
Li, S., Wang, X., Chen, Y., Tian, M., Lin, P., Lai, M., & Jiang, L. (2026) · Frontiers in Medicine · doi:10.3389/fmed.2026.1810098 · Cited by 4
Chain of Evidence: Pixel-Level Visual Attribution for Iterative Retrieval-Augmented Generation
Liu, P., Cui, Z., Wang, X., Liang, D., & Ye, W. (2026) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3805712.3809540 · Cited by 5
Shao, C., Qin, X., Li, X., Yang, M., Yuan, D., & Zheng, S. (2026) · Automation in Construction · doi:10.1016/j.autcon.2026.107041 · Cited by 2
Evaluation of ChatGPT-5.2 responses to frequently asked questions about benign bone tumors
Ayhan, B., Yoğurt, S. B., & Ayhan, Z. D. (2026) · Frontiers in Oncology · doi:10.3389/fonc.2026.1826346 · Cited by 2
Large Language Models for Pharmaceutical Knowledge Management: A Critical Review
Torres, A., & Fernandez, M. (2025) · Pharmacophore · doi:10.51847/xphozuyfsy · Cited by 2
Constructing Multimodal Wireless Knowledge Graphs for Large Language Model–Based Network Reasoning
Lin, W., Song, Y., Liu, Y., & Li, Y. (2026) · Applied Artificial Intelligence Research · doi:10.65455/1bt00146 · Cited by 2
“During the process of text blocking, topological constraints, causal relationships, and temporal dependencies are often segmented or lost, which has been recognized as a fundamental limitation of text-centric RAG in structured domains [36] .”
Ethical Considerations in Personal Health Large Language Models
Liu, J., & Liu, S. (2026) · Journal of Medical Internet Research · doi:10.2196/92240 · Cited by 2
“PH-LLMs should incorporate a sequential safety pipeline that combines input filtering, retrieval-augmented generation from curated authoritative sources, output checking against structured medical knowledge resources, and uncertainty signaling where technically feasible [85,116-118].”
A roadmap for medical large language models: a review of foundations, applications, and challenges
Sha, Y.-Y., Yu, L., Lin, Z.-H., Kaur, A., Lou, Y., Gornale, S. S., Zhang, T., Wong, L. S., Wang, Z., Yan, Y., et al. (2026) · Military Medical Research · doi:10.1016/j.mmr.2026.100050 · Cited by 4
“Collectively, these approaches aim to reduce domain mismatch, guide clinically appropriate model behavior, enable resource-aware customization, and support knowledge grounding through external evidence [91,103,104].”
InfluentialRAG in clinical practice: a cautionary tale of AI ‘Truthfulness’
Jung, H., CHO, K., Jun, T. J., & Kim, Y. (2026) · npj Health Systems · doi:10.1038/s44401-026-00115-x · Cited by 3
“RAG has emerged as a key architectural solution 5 , designed to mitigate this risk by grounding model outputs in curated, evidence-based knowledge sources 6 – 9 .”
Leveraging Large Language Models in Healthcare: From Speech Documentation to Conversational Agents
Winkler, C. (2026) · Springer eBooks · doi:10.1007/978-3-032-16837-5_16 · Cited by 2
Takamura, T., & Umezawa, A. (2025) · medRxiv · doi:10.1101/2025.10.20.25337146 · Cited by 3
“These types of queries may yield ambiguous responses without clear contextual or normative guidance, requiring particular caution in future applications (Amugongo et al., 2025; Gargari & Habibi, 2025).”
Ye, S., Kong, X., He, X., Yan, G., & Oh, D. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.03476 · Cited by 3
“Medical RAG system surveys confirm that external knowledge sources are an important path for mitigating hallucinations and enhancing transparency [32].”
When Confidence Takes the Wrong Path: Diagnosing Retrieval-State Lock-In in RAGPreprint
Julka, S. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.22728 · Cited by 2
Brooks, J., Blankson, P., Campbell, P., Cowley, R. A., Yang, T. Y., Oseni, T. I. A., Rodriguez, A., & Idris, M. Y. (2025) · JMIR Formative Research · doi:10.2196/82487 · Cited by 1
“Methodological advancements, such as instruction tuning, retrieval-augmented generation, and multistep reasoning, have demonstrated significant improvements in both diagnostic accuracy and clinical utility of LLMs [3-5].”
Wang, E. T., Song, S., Peng, K. C., & Liu, T. Y. A. (2026) · Asia-Pacific Journal of Ophthalmology · doi:10.1016/j.apjo.2026.100276 · Cited by 1
The Research on Intelligent Medical Triage System Based on Large Language Models
Zhao, X. (2025) · IEEE International Conference on Bioinformatics and Biomedicine (BIBM) · doi:10.1109/bibm66473.2025.11356579 · Cited by 1
“Lameck Mbangula Amugongo’s research team (2025) elaborated on the development stages of three RAG approaches in the medical field: naive RAG, advanced RAG, and modular RAG [5].”
Wulf, J., Meierhofer, J., & Pascher, N. (2026) · Procedia CIRP · doi:10.1016/j.procir.2025.09.030 · Cited by 1
Waldock, W. J., Gilchrist, M. J., Ashrafian, H., Darzi, A. W., & Franklin, B. D. (2026) · npj Antimicrobials and Resistance · doi:10.1038/s44259-026-00187-7 · Cited by 2
“Retrieval-augmented generation (RAG) addresses these issues by coupling LLMs with access to curated, evidence-based guidelines, thereby enhancing both factual reliability and clinical alignment 5 .”
Evaluation of LLMs in Retrieving Food and Nutritional Context for RAG Systems
Vavken, M. P., Ogrinc, M., Eftimov, T., & Seljak, B. K. (2025) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata66926.2025.11401545 · Cited by 1
“A comprehensive review by Amugongo et al. (2025) [6] assessed RAG-based approaches employed by LLMs in health-care, focusing on the different steps of retrieval, augmentation, and generation.”
Li, H., Cao, J., Duan, S., Hu, S., Lou, L., Lin, M., Jian, T., Shao, J., Zhang, X., Chen, P., et al. (2026) · npj Digital Medicine · doi:10.1038/s41746-026-02510-y · Cited by 1
“By incorporating a retrieval step, RAG could anchor to veri fi able information to reduce hallucinations and enhance the transparency of the sources from which the LLM ’ sresponseisgenerated 45,46 .”
Natural language querying of biological databases with large language models
Makarov, V. A., Stroganov, O. V., Furlong, L. I., Evarts, B., van den Biggelaar, L., Goulas, A., Stolte, E., Marren, D., & Greiffenberg, L. (2026) · Drug Discovery Today · doi:10.1016/j.drudis.2026.104654 · Cited by 1
Comparative Evaluation of Large Language Models in Clinical Diagnostics for Real-World Medical Cases
Młodawski, M., Deniziak, S. R., Płaza, M., Kwiatkowska, A., Jaszczyk, D., Gajda, A., & Twardowski, K. (2026) · Applied Sciences · doi:10.3390/app16073499 · Cited by 1
Hariharan, S., Balapriya, S., Jose, P. S. H., Xu, Q., Lu, J., & Kor, A.-L. (2026) · Expert Systems with Applications · doi:10.1016/j.eswa.2026.132388 · Cited by 2
TheraMind AI: A Multilingual and Multidomain Diagnostic Assistant for Psychotherapy
Reddy, J. A., Yasmeen, Goura, K., Reddy, Y. C. A. P., Rani, K. M. S., & Devi, P. D. Y. R. (2026) · International Conference on Innovative Practices in Technology and Management (ICIPTM) · doi:10.1109/iciptm69057.2026.11465737 · Cited by 1
Metadata
- Authors
- Amugongo, L. M., Mascheroni, P., Brooks, S., Doering, S., & Seidel, J.
- Affiliations
- Boehringer Ingelheim (Germany) (DE)
- Boehringer Ingelheim (China) (CN)
- Corresponding author
- Amugongo, Lameck Mbangula
- Venue
- PLOS Digital Health
- Publisher
- Public Library of Science (PLoS)
- Published
- 2025-06-11
- Volume
- 4
- Issue
- 6
- Pages
- e0000877
- ISSN
- 2767-3170 (ISSN-L)
- PMC
- PMC12157099
- OpenAlex
- W4411203672
- Semantic Scholar
- 5879575701b9b65b5cc56c00d9eebbfa219e0428
- Cited by
- 267
- References
- 126
- Citations by year
- 2026: 218 · 2025: 48
- FWCI
- 15.74
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- PLOS Digital Health (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Topics
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
- Keywords
- retrieval-augmented generation, large language models, health care, medical question answering, hallucinations, external knowledge sources, medical datasets
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.