Related Experiment Video
Updated: Jan 19, 2026
Digital Hybrid Model Preparation for Virtual Planning of Reconstructive Dentoalveolar Surgical Procedures
Published on: August 5, 2021
Applying deep artificial neural network approach to maxillofacial prostheses coloration
Yuichi Mine1, Shunsuke Suzuki2, Toru Eguchi3
1Department of Medical System Engineering, Division of Oral Health Sciences, Graduate School of Biomedical and Health Sciences, Hiroshima University, 1-2-3 Kasumi Minami-ku, Hiroshima 734-8553, Japan; Translational Research Center, Hiroshima University, 1-2-3 Kasumi Minami-ku, Hiroshima 734-8553, Japan.
Deep learning artificial neural networks (ANN) offer superior color matching for maxillofacial prostheses compared to random forest algorithms. This AI approach enhances prosthetic aesthetics and patient function.
Area of Science:
- Biomaterials science
- Artificial intelligence in medicine
- Prosthodontics
Background:
- Maxillofacial prostheses restore function and aesthetics after facial defects.
- Deep learning applications are expanding in medical fields.
- Accurate coloration is crucial for realistic prostheses.
Purpose of the Study:
- To apply an artificial neural network (ANN)-based deep learning approach for maxillofacial prosthesis coloration.
- To compare ANN-based deep learning with the random forest algorithm for pigment compounding accuracy.
Main Methods:
- Prepared 52 silicone elastomer specimens with varying colors.
- Measured CIE 1976 L* a* b* color space data using a spectrophotometer.
- Compared ANN and random forest algorithms for predicting pigment compounding amounts.
- Validated predictions by fabricating specimens and measuring color differences (CIEDE00 ΔE00).
Main Results:
- The ANN approach yielded a lower color difference (3.45 ± 0.87) compared to the random forest algorithm (5.54 ± 1.41).
- Deep ANN demonstrated superior accuracy in matching silicone elastomer colors to real skin tones.
- The ANN-based method achieved better color fidelity for maxillofacial prosthetic fabrication.
Conclusions:
- Deep artificial neural networks show significant promise for improving maxillofacial prosthesis coloration.
- This AI-driven technique can enhance the aesthetic outcomes of prosthetic rehabilitation.
- The findings suggest a valuable role for deep learning in creating more natural-looking prostheses.
