Generative Artificial Intelligence-Driven Clinical Case Simulation in Temporomandibular Disorder Education: ChatGPT...

Area of Science:

• Dental Education
• Artificial Intelligence in Medicine
• Diagnostic Reasoning

Background:

• Temporomandibular disorders (TMDs) and orofacial pain (OFP) require advanced diagnostic skills in dental education.
• Traditional patient simulations have limitations in availability and standardization.
• Generative artificial intelligence (GAI) presents a novel tool for clinical training.

Purpose of the Study:

• To compare the effectiveness of GAI (ChatGPT-3.5) versus standardized real patient simulations for training dental students in TMD/OFP diagnosis.
• To evaluate quantitative and qualitative aspects of student interactions with both simulation types.

Main Methods:

• A blinded, cross-sectional, crossover study involving 30 undergraduate dental students.
• Students completed two simulated cases: one with ChatGPT-3.5 and one with standardized real patients.
• Cases were validated by specialists; AI incorporated via structured prompts. Analysis included response metrics, information density, clarity, empathy, and diagnostic accuracy.

Main Results:

• GAI simulations demonstrated higher information density and clearer reasoning flow.
• Students required more follow-up questions and reformulations with real patients, indicating more adaptive communication.
• Diagnostic accuracy was comparable between GAI and real patient simulations; relevant information positively correlated with accuracy.

Conclusions:

• ChatGPT-3.5 provides comparable diagnostic reasoning to real patient simulations for TMD/OFP.
• Integrating GAI's scalability and standardization with real patient variability can enhance clinical competency training for dental professionals.