Related Experiment Video
Updated: May 23, 2025
10:42
A Postoperative Evaluation Guideline for Computer-Assisted Reconstruction of the Mandible
Published on: January 28, 2020
6.4K
Comparative Analysis of Artificial Intelligence Platforms in Generating Post-Operative Instructions for Rhinologic
Ariana L Shaari1, Shreya Bhalla2, Annie Xu1
1Department of Otolaryngology- Head and Neck Surgery, Rutgers New Jersey Medical School, Newark, New Jersey USA.
Summary
Artificial intelligence (AI) tools like ChatGPT and Gemini can create understandable, but not actionable, post-operative instructions for surgeries like endoscopic sinus surgery (ESS). Patients should use AI-generated medical information cautiously due to potential drawbacks.
Area of Science:
- Medical informatics
- Artificial intelligence in healthcare
- Patient education
Background:
- Rapid advancements in AI platforms like ChatGPT and Google Gemini.
- Increasing patient reliance on AI for medical information.
- The critical need for clear post-operative instructions for patient safety and optimal outcomes.
Purpose of the Study:
- To evaluate the accuracy and understandability of post-operative instructions generated by ChatGPT and Gemini.
- To compare the performance of ChatGPT and Gemini in creating patient-facing surgical information.
Main Methods:
- Generated post-operative instructions for endoscopic sinus surgery (ESS), balloon sinuplasty, Caldwell Luc surgery, and septoplasty using ChatGPT and Gemini.
- Assessed readability using standard metrics.
- Utilized the Patient Education Materials Assessment Tool (PEMAT) to evaluate understandability and actionability.
Main Results:
- Both ChatGPT and Gemini produced understandable post-operative instructions.
- Instructions generated by both AI platforms were found to be poor in terms of actionability.
- Readability scores indicated limitations in the clarity and ease of understanding for patient action.
Conclusions:
- AI platforms show potential for generating accessible medical information.
- Current AI-generated post-operative instructions lack sufficient actionability and optimal readability.
- Patients must exercise caution and be aware of the limitations of AI when seeking post-operative guidance.
