JMIR Publications

ABSTRACT

Background:

Medical education challenges students to manage vast amounts of complex information. Traditional mnemonic resources often follow a standardized approach, which may not accommodate diverse learning styles.

Objective:

This study aimed to develop a method for personalizing mnemonic generation using artificial intelligence (AI) and prompt engineering.

Methods:

We developed Personalized Multimodal Mnemonics (PMMs) using ChatGPT (GPT-4 model) for text mnemonic generation and DALL-E 3 for visual mnemonic creation. The process involved: (1) selecting medical concepts relevant to the current curriculum, (2) developing and refining prompt engineering principles, (3) generating text mnemonics, (4) creating corresponding visual representations, (5) integrating text and visual elements, and (6) personalizing mnemonics based on individual preferences.

Results:

The study generated PMMs for six medical concepts. Text mnemonics captured key information using various techniques, while visual representations provided complementary imagery. The iterative prompt engineering process contributed to refining outputs and ensuring relevance. Limitations included occasional inaccuracies in visual representations and text within images.

Conclusions:

This study presents an AI-assisted approach to mnemonic generation in medical education. The Personalized Multimodal Mnemonics (PMM) method integrates visual and verbal elements tailored to individual learning preferences. By leveraging AI's capabilities, this approach offers a pathway to more engaging and personalized learning experiences for medical students, while fostering digital literacy skills essential for future healthcare professionals. Additionally, this method eliminates the need for developing a separate mnemonics database, streamlining the learning process.