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

Area of Science:

• Medical Informatics
• Artificial Intelligence in Medicine
• Bibliometrics

Background:

• Large language model (LLM)-based AI chatbots are increasingly utilized in various fields, including surgery.
• Specific applications and research trends of these AI chatbots in surgical disciplines are not well-documented.

Purpose of the Study:

• To conduct a bibliometric analysis of research trends and potential applications of LLM-based AI chatbots in surgery.
• To identify key research areas, leading institutions, prolific authors, and emerging application domains.

Main Methods:

• Bibliometric analysis using Web of Science Core Collection data.
• Data analysis performed with VOSviewer, CiteSpace, and the R package bibliometrix.
• Inclusion criteria applied to 1372 initial papers, resulting in 260 relevant publications.

Main Results:

• Significant increase in research output from 2023 to 2024, led by the United States (52.1%).
• Harvard Medical School is the leading institution; key authors identified include Seth Ishith, Lechien Jerome, Cho Samuel, and Zaidat Bashar.
• Top keywords are "artificial intelligence" and "ChatGPT"; primary applications are in otolaryngology, plastic surgery, neurosurgery, and bariatric surgery, focusing on education.

Conclusions:

• AI chatbots demonstrate significant potential for enhancing patient and medical education within surgical fields.
• Challenges remain in developing sophisticated AI chatbots for accurate healthcare interactions and personalized patient care.
• This study provides a landscape overview and identifies future research directions for AI in surgery.