Exploring Radiology Postgraduate Students' Engagement with Large Language Models for Educational Purposes: A Study of...

Area of Science:

• Medical Education
• Artificial Intelligence in Medicine
• Radiology Training

Background:

• Large language models (LLMs) are increasingly explored for enhancing medical education.
• Limited understanding exists regarding radiology postgraduate students' engagement with LLMs.
• This study investigates their knowledge, attitudes, and practices concerning LLMs in education.

Purpose of the Study:

• To assess radiology postgraduate students' familiarity with LLMs.
• To evaluate their attitudes towards incorporating LLMs in medical education.
• To understand their current practices in utilizing LLMs for learning.

Main Methods:

• A cross-sectional quantitative survey was administered online.
• Participants were recruited nationwide using social media and snowball sampling.
• A validated questionnaire assessed knowledge, attitude, and practice regarding LLMs.

Main Results:

• 252 radiology postgraduate students participated, with 47.62% familiar with LLMs.
• A majority (71.82%) see potential in integrating LLMs with traditional tools.
• Students show a positive attitude (mean score 3.75) but lower practice (mean score 3.15) due to limitations.

Conclusions:

• Radiology postgraduate students are aware of LLMs and acknowledge their potential educational value.
• Positive attitudes exist, but practical application is limited by concerns over LLM limitations.
• Further integration strategies should address these concerns to optimize LLM use in radiology education.