A bibliometric analysis of large language model-based AI chatbots in surgery

Large language model-based artificial intelligence (AI) chatbots are gaining traction in surgery, yet their specific applications remain understudied. This bibliometric analysis assesses research trends and potential applications of large language model-based AI chatbots in surgery. We conducted a search in the Web of Science Core Collection database and analyzed the data using VOSviewer, CiteSpace, and the R package bibliometrix. Out of an initial 1372 papers, 260 met the inclusion criteria. Research output has significantly increased from 2023 to 2024. The United States led in publications, accounting for 52.1% of the total. Harvard Medical School emerged as the leading institution with 13 relevant publications, representing 5% of the overall output. The field comprises 1418 authors, with Seth Ishith, Lechien Jerome, Cho Samuel, and Zaidat Bashar being the most prolific, while Gupta Rohun is the most frequently co-cited author. The analysis of the top 20 keywords reveals that “artificial intelligence” and “ChatGPT” are the most common. Key application areas identified include otolaryngology – head and neck surgery, plastic surgery, neurosurgery, and bariatric surgery, with an emphasis on patient and medical education. AI chatbots in surgery show great potential for advancing patient and medical education. However, the development of sophisticated chatbots capable of facilitating accurate healthcare interactions and delivering personalized care remains a challenge. This analysis provides a comprehensive overview of the current research landscape and highlights areas for future investigation.