Identification of Key Nodes and Global Research Trends of artificial intelligence (AI)/ Large Language Model (LLM) in Medical Education: A Bibliometric Analysis(1986-2024) (Preprint)

Abstract

<sec> <title>BACKGROUND</title> In recent years, artificial intelligence (AI)/ Large Language Model(LLM) has significantly transformed the field of medical education, prompting extensive research. Many important bibliometric nodes in this emerging field have yet to be explored. </sec> <sec> <title>OBJECTIVE</title> This study aims to synthesize diverse publications to analyze the bibliometric attributes of this field, including landmarks, emerging topics, and development status. Also, the formation pattern of bibliometrics in emerging fields will be analyzed simultaneously. </sec> <sec> <title>METHODS</title> We utilized the Web of Science Core Collection to download literature on AI in medical education. Bibliometric analysis was performed using Citespace v.6.3.R1 software, which facilitated the analysis of publication volume, collaboration within the field, citation networks, and keyword analysis. Additionally, we employed the Bibliometrix package based on R for generating conceptual and thematic maps related to the topic. </sec> <sec> <title>RESULTS</title> A total of 547 publications were retrieved from the Web of Science Core Collection, covering the period from 1986 to 2024. The five leading countries in terms of publication volume were the United States, England, China, Canada, and India. The most prolific journals included JMIR Medical Education, BMC Medical Education, Cureus Journal of Medical Science, Medical Teacher, and Academic Medicine. The top institutions contributing to this body of work were the University of London, National University of Singapore, Harvard University, and Stanford University. Other important bibliometric characteristics, such as high-yield authors, highly cited authors, and frequently collaborating authors, were also identified. A citation co-citation network was established to determine the key knowledge base and potential pivotal literature in the field. Citespace software was utilized to identify clusters and bursts of high-frequency terms, highlighting current hotspots within the discipline. The Bibliometrix toolkit provided conceptual and thematic maps to assess the development status and trends in the field. </sec> <sec> <title>CONCLUSIONS</title> AI/LLM in medical education has emerged as a burgeoning field in recent years. JMIR Medical Education was identified as a key node based on its notable bibliometric characteristics. In the early stages of this emerging discipline, the journal's submission calls significantly influence the bibliometric features of the literature, thereby promoting field development. The discipline is currently in a developmental phase, lacking well-defined subfields. Topics such as "nursing education," "digital health," "medical exams," and "conversational agents" have garnered increasing interest over time. Research related to ChatGPT and large language models appears to occupy a central and influential position. Furthermore, medical ethics, medical training, and skills training are emerging focuses of current development and innovation, particularly in gene technology. However, this analysis indicates that there has been insufficient attention given to clinical reasoning, undergraduate education, and virtual reality in the context of AI/LLM in medical education. </sec>

Ähnliche Arbeiten