Related Experiment Video
Updated: Aug 12, 2025
09:10
Digital Hybrid Model Preparation for Virtual Planning of Reconstructive Dentoalveolar Surgical Procedures
Published on: August 5, 2021
1.8K
Computer-aided design and 3-dimensional artificial/convolutional neural network for digital partial dental crown
Taseef Hasan Farook1, Saif Ahmed2, Nafij Bin Jamayet3
1Adelaide Dental School, The University of Adelaide, Adelaide, South Australia, Australia.
Scientific Reports
|January 28, 2023
Summary
This study developed a 3D convolutional neural network (3D-CNN) to generate partial dental crowns (PDC) for restorative dentistry. The AI model shows promise as a proof-of-concept for predicting and generating dental prostheses using computer-aided design.
Area of Science:
- Dental Technology
- Artificial Intelligence in Dentistry
- Restorative Dentistry
Background:
- Traditional methods for creating dental prostheses can be time-consuming and may lack precision.
- Advancements in digital dentistry and artificial intelligence offer potential solutions for improving efficiency and accuracy in restorative procedures.
Purpose of the Study:
- To develop and validate a 3-dimensional convolutional neural network (3D-CNN) for the automated generation of partial dental crowns (PDC).
- To evaluate the effectiveness of intraoral scanners in capturing data for AI-driven dental prosthesis design.
Main Methods:
- Phase 1: Evaluated desktop laser and intraoral scanners for data acquisition, selecting intraoral scanners for superior volumetric detail.
- Phase 2: Synthesized 120 tooth preparations from intraoral scans and had clinicians design PDCs using computer-aided design (CAD).
- Phase 3: Trained a 3D-CNN model using the most accurate PDCs (n=30) to predict and generate prostheses.
Main Results:
- Intraoral scanners provided greater volumetric detail compared to desktop laser scanners.
- Significant differences were observed in surface area, volume, and spatial overlap among clinician-designed PDCs.
- The 3D-CNN model achieved a validation accuracy of 60%, with sensitivity of 1.00 and precision ranging from 0.50-0.83.
Conclusions:
- A 3D-CNN can be trained to predict and generate partial dental crowns (PDC) for restorative dentistry applications.
- This study serves as a proof-of-concept for utilizing AI in computer-aided design (CAD) workflows for dental prostheses.
- Further research and development are warranted to optimize the AI model for clinical use.
