Integrating AI Transcription Software in Medical Education: A Comparative Study of OSCE Documentation

Area of Science:

• Medical Education
• Artificial Intelligence in Healthcare
• Clinical Documentation

Background:

• AI is increasingly integrated into clinical workflows.
• Medical education needs to adapt to AI-assisted documentation.
• Freed AI offers potential efficiency and quality improvements.

Purpose of the Study:

• Compare Freed AI-generated notes with student-generated notes.
• Evaluate AI's performance in simulated clinical encounters.
• Assess documentation quality across different medical conditions.

Main Methods:

• First-year medical students used Freed AI during Objective Structured Clinical Examinations (OSCEs).
• Encounters covered cough, falls, and back pain scenarios.
• Notes were scored for completeness, accuracy, and medical relevance; analyzed with two-way ANOVA.

Main Results:

• Freed AI significantly outperformed student notes (mean score advantage ~3.78).
• A significant interaction between AI and condition was observed, with AI excelling in back pain.
• Back pain scenarios yielded the lowest overall scores regardless of source.

Conclusions:

• Freed AI enhances medical documentation quality, especially in complex cases like back pain.
• AI tools can augment student performance in medical education.
• Continued focus on clinical reasoning and case complexity is vital.