Related Experiment Video
Updated: Oct 18, 2025
Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization
Published on: February 23, 2024
Artificial Intelligence Model to Detect Real Contact Relationship between Mandibular Third Molars and Inferior
Tianer Zhu1, Daqian Chen2, Fuli Wu2
1Stomatology Hospital, School of Stomatology, Zhejiang University School of Medicine, Clinical Research Center for Oral Disease of Zhejiang Province, Key Laboratory of Oral Biomedical Research of Zhejiang Province, Cancer Center of Zhejiang University, Hangzhou 310006, China.
A new AI model, MM3-IANnet, accurately detects the contact between mandibular third molars and the inferior alveolar nerve on panoramic X-rays. This artificial intelligence approach assists dentists, reducing the need for costly CBCT scans.
Area of Science:
- Dentistry
- Radiology
- Artificial Intelligence
Background:
- Assessing the relationship between mandibular third molars (MM3s) and the inferior alveolar nerve (IAN) is crucial for preventing nerve injury during dental procedures.
- Panoramic radiographs are commonly used but often lack the detail to definitively determine MM3-IAN contact, frequently necessitating cone-beam computed tomography (CBCT).
Purpose of the Study:
- To develop and evaluate a novel deep learning-based detection model (MM3-IANnet) for automatically assessing the real contact relationship between MM3s and the IAN using panoramic radiographs.
- To minimize pseudo-contact interference and reduce the reliance on CBCT imaging.
Main Methods:
- A deep learning network, YOLOv4-based MM3-IANnet, was applied to panoramic radiographs for MM3-IAN contact detection.
- The ground truth for the real contact relationship was established using CBCT imaging.
- Performance was evaluated using accuracy metrics, comparing MM3-IANnet, human dentists, and a combined dentist-MM3-IANnet approach.
Main Results:
- The MM3-IANnet model achieved an average precision (AP) of 83.02% in detecting MM3-IAN contact.
- Dentists achieved an AP of 76.45%.
- The cooperative approach combining dentists and the MM3-IANnet model demonstrated the highest performance with an AP of 88.06%.
Conclusions:
- The MM3-IANnet detection model shows significant promise as an artificial intelligence tool to aid dentists in accurately assessing MM3-IAN contact from panoramic radiographs.
- This AI approach can potentially reduce the frequency of CBCT use, offering a more efficient diagnostic pathway.
