Related Experiment Video
Updated: Apr 28, 2026
07:32
Author Spotlight: 3D Movement Assessment of Maxillary Posterior Teeth in Clear Aligner Treatment
Published on: February 23, 2024
2.1K
Automatic dental crown generation with spatial constraint modeling
Golriz Hosseinimanesh1, Farida Cheriet1, Ammar Alsheghri2,3
1Polytechnique Montréal University, Canada.
Journal of Medical Imaging (Bellingham, Wash.)
|April 27, 2026
Summary
This study introduces an advanced deep learning framework for automated dental crown generation, significantly improving geometric precision and functional accuracy for direct clinical use. The new method ensures better fit and reduces the need for manual adjustments in dental laboratories.
Area of Science:
- Computer-aided design (CAD) in dentistry
- Artificial intelligence (AI) for medical applications
- Biomedical engineering and informatics
Background:
- Current deep learning methods for dental crown generation primarily focus on shape completion.
- Existing approaches lack explicit modeling of critical spatial relationships like margin lines and occlusal contacts.
- This results in generated crowns with insufficient spatial accuracy for direct clinical application, necessitating manual adjustments.
Purpose of the Study:
- To develop a comprehensive framework for automated dental crown mesh generation with enhanced geometric precision and functional accuracy.
- To address the limitations of current methods by integrating explicit spatial relationship modeling.
- To enable direct clinical use of AI-generated dental crowns, reducing laboratory design time.
Main Methods:
- A transformer encoder-decoder architecture integrated with differentiable Poisson surface reconstruction was employed.
- Margin line data was incorporated as direct network input, alongside master and antagonist arch geometries, for explicit boundary constraints.
- Spatial constraint losses, including antagonist interaction loss and intersection loss, were implemented to ensure anatomical validity and prevent interferences.
Main Results:
- The proposed framework demonstrated substantial geometric accuracy improvements (35.9%–40.6%) over state-of-the-art methods.
- Margin line integration improved geometric precision by 31.2%, reducing maximum boundary errors and variability.
- Antagonist interaction loss enhanced occlusal alignment by 9.51%, and intersection loss significantly minimized crown penetration into adjacent teeth.
Conclusions:
- The integration of spatial constraint modeling and direct margin line input significantly enhances automated dental crown generation.
- The framework achieves substantial performance improvements, validating its effectiveness for clinical deployment.
- This work establishes a foundation for advanced, automated dental crown design systems in clinical practice.
