Evaluation of Different Color Determination Methods

Overview

With the increasing aesthetic demands in dental practice, the accuracy and consistency of tooth color determination methods have gained importance for clinical success. This study aims to compare different tooth color determination methods in terms of accuracy and repeatability under in vivo conditions. A spectrophotometric system (VITA Easyshade V) will be used as the reference method. In comparison, the results obtained through visual assessment by observers, a digital intraoral scanner (TRIOS 3), and artificial intelligence-based color analysis systems will be evaluated. Statistical analyses will be conducted based on ΔE, L\*, a\*, b\*, and shade code data obtained from the measurements, and the agreement among the methods as well as their correlations with the spectrophotometer will be examined. This study aims to demonstrate the clinical validity of digital and AI-assisted methods and to emphasize the importance of objective approaches in tooth shade selection. The aim of this study is to compare four different tooth color determination methods used in aesthetic dentistry-spectrophotometric system, observer-based visual assessment, digital intraoral scanner, and artificial intelligence-assisted analysis system-in terms of accuracy and repeatability under in vivo conditions. The study seeks to statistically evaluate and compare the color measurement values obtained from these methods with those of the spectrophotometric system, which is considered the reference method, and to determine the level of agreement among the methods. In addition, it aims to assess the clinical applicability and reliability of digital technologies and artificial intelligence-based systems in tooth color selection.

This analytical cross-sectional clinical study was conducted between May 2025 and June 2025 at the Department of Prosthodontics, Faculty of Dentistry, Necmettin Erbakan University. The study aims to demonstrate the clinical validity of digital and artificial intelligence-assisted methods and to contribute to emphasizing the importance of objective approaches in tooth shade selection. A total of 40 adult participants with healthy maxillary central incisors were included in the study. Color measurements will be taken from the middle third of the labial surface of each participant's maxillary central incisor. All measurements will be performed by the same clinician, in a fixed position, and under standardized lighting conditions. Color Determination Methods: 1. Spectrophotometric Measurement: The VITA Easyshade V spectrophotometer will be used as the reference method. The device will be calibrated before each use, and measurements will be taken in a perpendicular position to the tooth surface. VITA 3D Master shade values as well as L\*, a\*, and b\* parameters will be recorded. 2. Visual Assessment: Observer-based visual evaluation will be performed using the VITA 3D Master shade guide. To minimize visual bias, the evaluation time will be limited to 10 seconds, and the selected shade values will be converted into the CIE Lab\* color system. 3. Digital Intraoral Scanner: The TRIOS 3 digital intraoral scanner will be used for image acquisition following device calibration. The color values obtained from the scans will be recorded. 4. Artificial Intelligence-Assisted Color Analysis: Photographs of the teeth will be taken in RAW format under natural light, from a fixed distance, and against a neutral gray background. After preprocessing, the images will be analyzed using a specially trained image-processing model based on artificial intelligence. Data Analysis: The color values obtained from each method will be compared with those from the spectrophotometric measurements. Color differences will be calculated using the ΔE formula. One-way ANOVA and Bonferroni post-hoc tests will be employed to evaluate differences among the methods. The agreement between L\*, a\*, and b\* values will be analyzed using Kendall's tau-b correlation analysis, while the consistency of shade code matching will be assessed using Cohen's Kappa statistics. All data will be processed and evaluated using an appropriate statistical analysis software program.