Background: The emergence of large language models has led to significant progress in various fields. In recent years, researchers have had a special focus on its applications in medical education.
Objectives: This study explores the social and intellectual framework of research on
large language models as a growing technology in medical education across the Web of Science database.
Methods: The present study is a quantitative study performed using scientometric techniques. The statistical population consisted of texts containing 643 documents from the year of this study. Following the Prisma guidelines, studies were screened based on their titles, abstracts, and keywords, leading to the selection of 233 documents for the research population. No restrictions (such as year, time, country) were applied to the search. Data analysis was undertaken using VOSviewer software.
Results: The study found that the United States was the most active country, generating 90 papers and receiving 537 citations. Author Lee, Hyunsoo, received 163 citations, with JMIR Medical Education being among the most frequent authors and journals with 30 papers and 352 citations. The keywords “artificial intelligence” with 144 repetitions, “ChatGPT” with 131 repetitions, “medical education” with 97 repetitions, and “large language models” with 89 repetitions make them highly frequent words. Four clusters were formed from the keywords obtained in the study”.
Conclusion: The research in large language models and medical education is growing rapidly, with organizations revealing a strong interest in international cooperation. This field has great potential for development. Nevertheless, countries with more publications have not necessarily received more citations.