Back to Search Start Over

Comparative analysis of optical and numerical models for reflectance and color prediction of monolithic dental resin composites with varying thicknesses.

Authors :
Tejada-Casado, Maria
Duveiller, Vincent
Ghinea, Razvan
Gautheron, Arthur
Clerc, Raphaël
Salomon, Jean-Pierre
Pérez, María del Mar
Hébert, Mathieu
Herrera, Luis Javier
Source :
Dental Materials. Oct2024, Vol. 40 Issue 10, p1677-1684. 8p.
Publication Year :
2024

Abstract

To assess the prediction accuracy of recent optical and numerical models for the spectral reflectance and color of monolithic samples of dental materials with different thicknesses. Samples of dental resin composites of Aura Easy Flow (Ae1, Ae3 and Ae4 shades) and Estelite Universal Flow Super Low (A1, A2, A3, A3.5, A4 and A5 shades) with thicknesses between 0.3 and 1.8 mm, as well as Estelite Universal Flow Medium (A2, A3, OA2 and OA3 shades) with thicknesses between 0.4 and 2.0 mm, were used. Spectral reflectance and transmittance factors of all samples were measured using a X-Rite Color i7 spectrophotometer. Four analytical optical models (2 two-flux models and 2 four-flux models) and two numerical models (PCA-based and L*a*b*-based) were implemented to predict spectral reflectance of all samples and then convert them into CIE-L*a*b* color coordinates (D65 illuminant, 2°Observer). The CIEDE2000 total color difference formula (Δ E 00 ) between predicted and measured colors, and the corresponding 50:50% acceptability and perceptibility thresholds (A T 00 and P T 00 ) were used for performance assessment. The best performing optical model was the four-flux model RTE-4F-RT, with an average Δ E 00 = 0.72 over all samples, 94.87% of the differences below A T 00 and 65.38% below P T 00 . The best performing numerical model was L*a*b*-PCHIP (interpolation mode), with an average Δ E 00 = 0.48, and 100% and 79.69% of the differences below A T 00 and P T 00 , respectively. Both optical and numerical models offer comparable color prediction accuracy, offering flexibility in model choice. These results help guide decision-making on prediction methods by clarifying their strengths and limitations. • Reflectance data of dental materials can be estimated by numerical algorithms. • Reflectance data of dental materials can be estimated by optical models. • Predicted — measured color differences are generally lower than AT 00. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
01095641
Volume :
40
Issue :
10
Database :
Academic Search Index
Journal :
Dental Materials
Publication Type :
Academic Journal
Accession number :
179632018
Full Text :
https://doi.org/10.1016/j.dental.2024.07.013