Related Experiment Video
Updated: May 7, 2026
09:10
Digital Hybrid Model Preparation for Virtual Planning of Reconstructive Dentoalveolar Surgical Procedures
Published on: August 5, 2021
1.9K
Toward artificial intelligence in dental prosthesis planning - a preliminary in-silico feasibility study
Michael Del Hougne1, Philipp Del Hougne2, Isabella Di Lorenzo3
1Department of Prosthodontics, University of Würzburg, Pleicherwall 2, Würzburg, 97070, Germany. hougne_m@ukw.de.
BMC Oral Health
|August 31, 2025
Summary
An artificial neural network (ANN) accurately learned dental prosthesis planning, achieving 99.51% accuracy. This technology shows potential to assist clinicians and improve patient self-assessment in prosthodontics.
Area of Science:
- Artificial Intelligence in Dentistry
- Machine Learning for Clinical Decision Support
Background:
- Dental prosthesis planning is complex, requiring individualized treatment based on findings and guidelines.
- Assessing the potential of artificial neural networks (ANNs) to replicate this planning process was the study's objective.
Purpose of the Study:
- To evaluate if an ANN can accurately approximate dental prosthesis planning.
- To determine the impact of training data, ANN architecture, and initialization on performance.
Main Methods:
- Dental prosthesis planning was modeled as a multi-output, multi-class classification problem.
- An ANN was trained using supervised learning on synthetic datasets.
- Performance was evaluated on unseen data, varying ANN parameters.
Main Results:
- The ANN achieved a high accuracy of 99.51% (±0.15) in approximating dental prosthesis planning.
- Results were consistent across different ANN initializations, training set sizes, and architectures.
Conclusions:
- ANNs can effectively learn dental prosthesis planning, as demonstrated in this in-silico study.
- ANNs show potential to support clinicians with automated recommendations, enhancing decision-making.
- This technology may enable patient self-assessment and improve prosthodontic care.
