Intelligent virtual case learning system based on real medical records and natural language processing.

Area of Science:

• Medical Education Technology
• Artificial Intelligence in Healthcare
• Clinical Reasoning Development

Background:

• Modernizing medical education with AI and technology is crucial for improving students' clinical thinking.
• Online learning is vital due to resource shortages, social distancing needs, and rising education costs.
• Virtual case learning systems using NLP can enhance clinical thinking by processing real medical records.

Purpose of the Study:

• To develop AIteach, a virtual case system utilizing hospital records and NLP.
• To improve clinical thinking abilities through a contactless, self-service, trial-and-error learning application.

Main Methods:

• Case extraction from hospital data centers using NLP techniques like word segmentation and synonym conversion.
• Establishment of modules for clinical questioning, virtual cases, and student learning difficulties for simulation.
• Assessment of 15 graduate students using pre- and post-learning simulation tests with multi-dimensional evaluation indicators.

Main Results:

• The AIteach system employs a double-cycle learning model for case information collection and analysis.
• Student performance showed a significant average score increase from 69.87 to 85.6 (P < 0.01).
• Key clinical thinking indicators improved, with a notable 47% enhancement in logical reasoning.

Conclusions:

• AIteach offers an online tool for clinical thinking training for medical students, integrating real cases and NLP.
• Virtual case learning facilitates clinical thinking development, especially when in-person tutoring is limited by circumstances like pandemics.