Panoramic Landmarks: Comparing LLM-Assisted, Manual Tracing, and Self-Directed Learning in Dental Education

Area of Science:

• Dental Radiology Education
• Anatomical Landmark Identification

Background:

• Accurate identification of anatomical landmarks on panoramic radiographs is crucial in dentistry but challenging.
• Traditional teaching methods often need supplementary tools for skill development.
• This study compares self-directed learning (SDL), manual tracing (MT), and AI (ChatGPT) for learning radiographic landmarks.

Purpose of the Study:

• To evaluate and compare the efficacy of three supplementary learning modalities for identifying anatomical landmarks on panoramic radiographs.
• To determine the effectiveness of self-directed learning, manual tracing, and AI-assisted learning in dental radiology education.
• To assess the impact of different learning methods on dental students' ability to identify specific radiographic landmarks.

Main Methods:

• A prospective study involving 63 third-year dental students divided into three groups (SDL, MT, ChatGPT).
• Students received a lecture and were tested immediately (baseline) and at a 4-week follow-up.
• Intra- and intergroup differences were analyzed using non-parametric statistical tests.

Main Results:

• The manual tracing (MT) group achieved significantly higher overall scores compared to SDL and ChatGPT groups.
• Significant improvements from baseline were observed in the MT group for 24 landmarks and in the ChatGPT group for 16 landmarks.
• The ChatGPT group outperformed MT in identifying four specific landmarks, including the zygomatic process.

Conclusions:

• Manual tracing (MT) is the most effective overall method for learning anatomical landmarks on panoramic radiographs.
• ChatGPT demonstrated value in learning specific landmarks, suggesting a supplementary role for AI.
• An integrated approach combining traditional manual tracing with AI-assisted learning is recommended to enhance dental radiology education.