Objective: This study sought to explore the unexamined capabilities of ChatGPT in describing the surgical steps of a specialized operation, the Fisher cleft lip repair.
Design: A chat log within ChatGPT was created to generate the procedural steps of a cleft lip repair utilizing the Fisher technique. A board certified craniomaxillofacial (CMF) surgeon then wrote the Fisher repair in his own words blinded to the ChatGPT response. Using both responses, a voluntary survey questionnaire was distributed to residents of plastic and reconstructive surgery (PRS), general surgery (GS), internal medicine (IM), and medical students at our institution in a blinded study.
Setting: Authors collected information from residents (PRS, GS, IM) and medical students at one institution.
Main outcome measures: Primary outcome measures included understanding, preference, and author identification of the procedural prompts.
Results: Results show PRS residents were able to detect more inaccuracies of the ChatGPT response as well as prefer the CMF surgeon's prompt in performing the surgery. Residents with less expertise in the procedure not only failed to detect who wrote what procedure, but preferred the ChatGPT response in explaining the concept and chose it to perform the surgery.
Conclusions: In applications to surgical education, ChatGPT was found to be effective in generating easy to understand procedural steps that can be followed by medical personnel of all specialties. However, it does not have expert capabilities to provide the minute detail of measurements and specific anatomy required to perform medical procedures.
Keywords: artificial intelligence; craniofacial surgery; surgical technique.