Artificial Intelligence-Generated Podcasts for Wilderness Medicine Education: A Feasibility Study

Area of Science:

• Medical Education
• Artificial Intelligence in Healthcare
• Biomedical Informatics

Background:

• Podcasts are a valuable tool for medical education, particularly for reviewing journal articles.
• However, podcast production is resource-intensive, limiting their widespread use.
• Exploring AI-driven solutions can enhance accessibility and efficiency in medical education.

Purpose of the Study:

• To assess the feasibility of using a trained AI model to generate educational podcasts from journal articles.
• To evaluate listener perception and comprehension of AI-generated medical education podcasts.
• To determine if AI-generated content can effectively replace traditional resource-intensive podcast production.

Main Methods:

• Google Gemini 2.0 Flash AI was trained on 12 Wilderness and Environmental Medicine journal articles.
• Three distinct educational podcasts were generated using AI, featuring AI-generated host voices.
• Participants listened to AI-generated podcasts and completed perception surveys and knowledge assessments.

Main Results:

• Thirty-one participants completed the study, with highly positive feedback.
• 87.5% found the content accurate and relevant (mean Likert score 4.37).
• 81.25% felt the podcasts aided learning (mean Likert score 4.14), with a mean knowledge score of 88.7%.

Conclusions:

• AI-generated podcasts are a feasible and effective tool for medical education.
• These podcasts are positively received by learners and facilitate knowledge acquisition.
• AI offers a potential solution to resource limitations in creating educational audio content.