Related Experiment Video
Updated: Jul 31, 2025
06:28
E-Patient Counseling Trial E-PACO: Computer Based Education versus Nurse Counseling for Patients to Prepare for Colonoscopy
Published on: August 1, 2019
8.4K
Performance of ChatGPT on the Plastic Surgery Inservice Training Examination
Aesthetic Surgery Journal
|May 2, 2023
Summary
ChatGPT demonstrated 54.96% accuracy on the Plastic Surgery Inservice Training Examination (PSITE), indicating potential for AI in resident education. Further research is needed to explore its role in plastic surgery training.
Area of Science:
- Medical Education
- Artificial Intelligence in Medicine
- Plastic Surgery Training
Background:
- The Plastic Surgery Inservice Training Examination (PSITE) is a standardized tool for resident self-evaluation in Plastic Surgery programs.
- Large Language Models (LLMs) like ChatGPT show promise for advancing the field of Plastic Surgery.
Purpose of the Study:
- To evaluate the accuracy of ChatGPT as an educational tool for Plastic Surgery residents.
- To assess ChatGPT's performance on the 2022 PSITE.
Main Methods:
- Utilized 242 questions from the 2022 PSITE.
- Flagged and excluded image- or table-based questions.
- Analyzed ChatGPT responses for accuracy, categorizing errors (logical, informational, explicit fallacy).
Main Results:
- ChatGPT achieved an overall accuracy of 54.96% on the PSITE.
- Responses demonstrated high rates of logical reasoning (88.8%), internal information use (95.5%), and external information use (92.1%).
- A statistically significant difference was observed in ChatGPT's use of external information between correct and incorrect responses (P < .05).
Conclusions:
- ChatGPT has the potential to be a valuable tool in Plastic Surgery resident education.
- Its capabilities include providing general knowledge, clarifying information, facilitating case-based learning, and promoting evidence-based medicine.
- Advancements in AI and LLMs suggest a significant future role for tools like ChatGPT in surgical training.
