Exploring Generative Artificial Intelligence-Assisted Medical Education: Assessing Case-Based Learning for Medical...

Area of Science:

• Medical Education Technology
• Artificial Intelligence in Healthcare

Background:

• Generative Artificial Intelligence (GenAI) is increasingly accessible for medical education.
• Determining the optimal role of GenAI in medical school curricula is crucial for students and faculty.
• Understanding GenAI's capabilities and limitations is essential for effective integration.

Purpose of the Study:

• To assess the capabilities of a GenAI, specifically ChatGPT, in a pre-clerkship medical curriculum.
• To evaluate GenAI's performance in generating educational materials, assessment questions, and answering medical queries.
• To explore GenAI's role in clinical reasoning through problem-based learning scenarios.

Main Methods:

• Evaluated ChatGPT's performance in generating educational content.
• Assessed its ability to create medical assessment questions.
• Tested its responses to medical queries and clinical reasoning scenarios within a case-based active learning framework.

Main Results:

• GenAI effectively addressed questions on epidemiology, diagnosis, and treatment.
• Inconsistent answer comprehensiveness was observed; critical details were sometimes omitted.
• GenAI struggled with complex topics, posing a risk of incomplete information for students.

Conclusions:

• GenAI can be a valuable tool in medical education but requires careful oversight.
• Faculty awareness of GenAI's strengths and weaknesses is vital for guiding student usage.
• Educating students on the limitations of GenAI is imperative to ensure accurate information retrieval and prevent knowledge gaps.