Image generation technology for functional occlusal pits and fissures based on a conditional generative adversarial...

Area of Science:

• Biomedical Engineering
• Computer Vision
• Dental Technology

Background:

• Natural teeth possess complex occlusal surface features, including pits and fissures, crucial for proper function.
• Existing dental restoration methods often lack detailed representation of these localized crown surface features.

Purpose of the Study:

• To propose and evaluate an image generation technology for creating accurate functional occlusal pits and fissures.
• To address the limitations in capturing detailed local crown surface morphology in current dental restoration techniques.

Main Methods:

• Construction of tooth depth image datasets using an orthogonal projection method.
• Optimization of model parameters incorporating jaw position spatial constraints, L1 loss, and perceptual loss functions.
• Evaluation of generated image quality using two metrics and deformation of dental crowns against expert data.

Main Results:

• The developed network generated high-quality images effectively restoring detailed pit and fissure features on dental crowns.
• Restored features exhibited a low standard deviation of 0.1802mm when compared to expert-designed tooth crown models.

Conclusions:

• The proposed image generation technology successfully restores complex occlusal pit and fissure morphology.
• This advancement offers improved precision for dental restorations by accurately recreating natural tooth surface details.