Enhancing wisdom teeth detection in panoramic radiographs using multi-channel convolutional neural network with...

Area of Science:

• Artificial Intelligence in Dentistry
• Medical Imaging Analysis
• Oral and Maxillofacial Surgery

Background:

• Panoramic radiographs are crucial for assessing wisdom teeth.
• Accurate detection of wisdom teeth aids in diagnosis and treatment planning.
• Current methods may involve complex annotation or subjective interpretation.

Purpose of the Study:

• To develop and evaluate a novel AI-based approach for detecting wisdom teeth in panoramic radiographs.
• To create a simplified and efficient pipeline for wisdom teeth detection.
• To enhance diagnostic consistency and support clinical decision-making.

Main Methods:

• A multi-channel convolutional neural network (CNN) was designed for wisdom teeth detection.
• A curated dataset of annotated panoramic dental images was utilized.
• Images were preprocessed into three channels (full, left, right) to mimic diagnostic workflow.
• A classification-based CNN model predicted wisdom teeth presence in four quadrants.

Main Results:

• The AI model achieved an accuracy of 82.46%.
• The area under the receiver operating characteristic curve (AUROC) was 0.8866.
• The area under the precision-recall curve (AUPRC) was 0.8542, indicating reliable performance.
• The method demonstrated robustness across diverse image conditions.

Conclusions:

• The proposed AI system offers a reliable and efficient method for detecting wisdom teeth.
• This approach reduces annotation overhead compared to traditional segmentation methods.
• The system can improve diagnostic objectivity and efficiency, especially for less experienced practitioners.