Empowering medical systematic reviews with large language models: methods, development directions, and applications

With the exponential growth of biomedical literature, traditional keyword-based retrieval methods are increasingly inadequate for meeting the dual demands of efficiency and precision in clinical and research contexts. In recent years, large language models (LLMs), exemplified by ChatGPT and DeepSeek...

Full description

Saved in:
Bibliographic Details
Main Authors: Yannan HUANG, Haoran SANG, Yu LIU, Liantao MA, Yinghao ZHU
Format: Article
Language:zho
Published: Editorial Office of Journal of Guangxi Medical University 2025-06-01
Series:Guangxi Yike Daxue xuebao
Subjects:
Online Access:https://journal.gxmu.edu.cn/article/doi/10.16190/j.cnki.45-1211/r.2025.03.001
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849426539993628672
author Yannan HUANG
Haoran SANG
Yu LIU
Liantao MA
Yinghao ZHU
author_facet Yannan HUANG
Haoran SANG
Yu LIU
Liantao MA
Yinghao ZHU
author_sort Yannan HUANG
collection DOAJ
description With the exponential growth of biomedical literature, traditional keyword-based retrieval methods are increasingly inadequate for meeting the dual demands of efficiency and precision in clinical and research contexts. In recent years, large language models (LLMs), exemplified by ChatGPT and DeepSeek, have demonstrated significant potential in supporting medical systematic reviews due to their powerful natural language processing capabilities. However, its inherent challenges such as the"hallucination"problem and lagging knowledge update limit the reliability of its direct application. This paper systematically introduces six core technical approaches currently used to mitigate hallucinations in LLMs, with a particular focus on explaining the principles and application advantages of retrieval-augmented generation (RAG). After comprehensively reviewing the technical characteristics and application scenarios of 22 representative studies in the context of systematic reviews, the paper further identifies LLMs capable of"structured understanding and generation based on levels of evidence"as one of the key future directions. The goal is to provide systematic guidance for medical researchers and clinical practitioners, helping them make scientific and efficient use of LLMs to enhance the efficiency of biomedical literature processing and the quality of evidence-based medical decision-making.
format Article
id doaj-art-2ff8d4c1552946fa9d8d4a53b9dbf68f
institution Kabale University
issn 1005-930X
language zho
publishDate 2025-06-01
publisher Editorial Office of Journal of Guangxi Medical University
record_format Article
series Guangxi Yike Daxue xuebao
spelling doaj-art-2ff8d4c1552946fa9d8d4a53b9dbf68f2025-08-20T03:29:21ZzhoEditorial Office of Journal of Guangxi Medical UniversityGuangxi Yike Daxue xuebao1005-930X2025-06-0142332333110.16190/j.cnki.45-1211/r.2025.03.001gxykdxxb-42-3-323Empowering medical systematic reviews with large language models: methods, development directions, and applicationsYannan HUANG0Haoran SANG1Yu LIU2Liantao MA3Yinghao ZHU4Peking University, Beijing 100871, ChinaPeking University, Beijing 100871, ChinaUniversity of Oxford, Oxford OX2 0JB, United KingdomPeking University, Beijing 100871, ChinaPeking University, Beijing 100871, ChinaWith the exponential growth of biomedical literature, traditional keyword-based retrieval methods are increasingly inadequate for meeting the dual demands of efficiency and precision in clinical and research contexts. In recent years, large language models (LLMs), exemplified by ChatGPT and DeepSeek, have demonstrated significant potential in supporting medical systematic reviews due to their powerful natural language processing capabilities. However, its inherent challenges such as the"hallucination"problem and lagging knowledge update limit the reliability of its direct application. This paper systematically introduces six core technical approaches currently used to mitigate hallucinations in LLMs, with a particular focus on explaining the principles and application advantages of retrieval-augmented generation (RAG). After comprehensively reviewing the technical characteristics and application scenarios of 22 representative studies in the context of systematic reviews, the paper further identifies LLMs capable of"structured understanding and generation based on levels of evidence"as one of the key future directions. The goal is to provide systematic guidance for medical researchers and clinical practitioners, helping them make scientific and efficient use of LLMs to enhance the efficiency of biomedical literature processing and the quality of evidence-based medical decision-making.https://journal.gxmu.edu.cn/article/doi/10.16190/j.cnki.45-1211/r.2025.03.001large language modelsmedical systematic reviewsretrieval-augmented generationprompt engineering
spellingShingle Yannan HUANG
Haoran SANG
Yu LIU
Liantao MA
Yinghao ZHU
Empowering medical systematic reviews with large language models: methods, development directions, and applications
Guangxi Yike Daxue xuebao
large language models
medical systematic reviews
retrieval-augmented generation
prompt engineering
title Empowering medical systematic reviews with large language models: methods, development directions, and applications
title_full Empowering medical systematic reviews with large language models: methods, development directions, and applications
title_fullStr Empowering medical systematic reviews with large language models: methods, development directions, and applications
title_full_unstemmed Empowering medical systematic reviews with large language models: methods, development directions, and applications
title_short Empowering medical systematic reviews with large language models: methods, development directions, and applications
title_sort empowering medical systematic reviews with large language models methods development directions and applications
topic large language models
medical systematic reviews
retrieval-augmented generation
prompt engineering
url https://journal.gxmu.edu.cn/article/doi/10.16190/j.cnki.45-1211/r.2025.03.001
work_keys_str_mv AT yannanhuang empoweringmedicalsystematicreviewswithlargelanguagemodelsmethodsdevelopmentdirectionsandapplications
AT haoransang empoweringmedicalsystematicreviewswithlargelanguagemodelsmethodsdevelopmentdirectionsandapplications
AT yuliu empoweringmedicalsystematicreviewswithlargelanguagemodelsmethodsdevelopmentdirectionsandapplications
AT liantaoma empoweringmedicalsystematicreviewswithlargelanguagemodelsmethodsdevelopmentdirectionsandapplications
AT yinghaozhu empoweringmedicalsystematicreviewswithlargelanguagemodelsmethodsdevelopmentdirectionsandapplications