Quantification of Color Variation of Various Esthetic Restorative Materials in Pediatric Dentistry
Pranshu Varshney,Saima Y Khan,Mahendra K Jindal,Yasser Azim,Aditi Bhardwaj,Vinod Kumar
DOI: https://doi.org/10.5005/jp-journals-10005-2905
Abstract:Aim of study: The goal of this paper is to find an association between the staining capacity of dental restorations used in pediatric patients and food items and to develop an optimum model to predict the most informative factor that causes the highest amount of color change through machine learning algorithms. Background: Color changes in restorative materials occur as a result of intrinsic and extrinsic factors, such as the type of restorative material, food items used, polished status of the material, and time interval. Materials and methods: This was an "in vitro study" conducted at Aligarh Muslim University, Aligarh, Uttar Pradesh, India. The study included 200 specimens, that is, 40 in each group A (orange juice), group B (Amul Kool Café), group C (Pepsi), group D (Amul Kesar Milk), and group E (artificial saliva). The materials were glass ionomer cement (GIC), resin-modified glass ionomer cement (RMGIC), microhybrid composite resin, and nanohybrid composite resin. These were further divided into polished and unpolished groups. The optimum modeling of the prediction of color change in materials by different effective factors was done by machine learning decision tree. We applied two algorithms: Chi-square automatic interaction detector (CHAID) and classification and regression tree (CART). In prediction modeling in the decision tree by CHAID and CART, color change is taken as the dependent variable, and group (type of restorative material), food items, time interval, and polished status are taken as independent variables. Results: The various beverages caused significant color variation due to different pigmentation agents. The agent that caused the highest color change was Kool Café. The Kesar Milk had the lowest pigmentation capacity. The greatest color variation was found on Glasionomer FX-II submerged in Pepsi and the least on Ivoclar Te-Econom Plus in Kesar Milk. The mean absolute error for the training dataset in the CART model and CHAID model is 0.379 and 0.332, and for the testing data set, it is 0.398 and 0.333, respectively. Therefore, the prediction of color change by the CHAID model is optimum, and we found that the restorative materials have a maximum predictor importance of 0.86 (86%), time interval 0.07 (7%), food items 0.04 (4%), and polished status has the least importance, that is, 0.03 (3%). Conclusion: The staining capacity of restorative material highly depends on the material itself, the initial time interval, and least on the food items used. Clinical significance: The clinical performance of dental restorations could be affected by various beverages consumed by children. This study thus provides important clinical insights into esthetic dentistry by offering valuable information on long-term color stability and the effect of polishing on common esthetic restorative materials used in pediatric dentistry. How to cite this article: Varshney P, Khan SY, Jindal MK, et al. Quantification of Color Variation of Various Esthetic Restorative Materials in Pediatric Dentistry. Int J Clin Pediatr Dent 2024;17(7):754-765.