Application of Convolutional Neural Networks in an Automatic Judgment System for Tooth Impaction Based on Dental...

Area of Science:

• Artificial Intelligence in Dentistry
• Medical Image Analysis
• Dental Diagnostics

Background:

• Panoramic radiography (PANO) is essential for dental examinations, aiding diagnosis and treatment planning.
• Artificial intelligence (AI) integration offers potential for enhanced medical applications.
• Accurate detection of impacted teeth is crucial for effective dental care.

Purpose of the Study:

• To develop an automated system for impacted third molar segmentation and localization in PANO images.
• To enhance feature extraction using Sobel edge detection and image preprocessing techniques.
• To improve the accuracy and efficiency of impacted tooth detection using AI.

Main Methods:

• Implementing Sobel edge detection and enhancement for improved feature extraction.
• Training a convolutional neural network (CNN) for automated impacted tooth detection.
• Utilizing AI for segmentation and localization of impacted third molars in dental radiographs.

Main Results:

• The AI system achieved a detection accuracy of 98.66% after image preprocessing and enhancement.
• This represents a significant improvement over existing methods with approximately 90% accuracy.
• The automated detection process is rapid, completing in 4.4 seconds per image.

Conclusions:

• The developed AI system effectively supports clinical decision-making for dentists.
• Enhanced image preprocessing and CNNs significantly improve impacted tooth detection accuracy.
• This technology allows dental professionals to focus more on patient care and treatment planning.