Predicting Esthetic Alteration of Resin-Matrix Ceramics Under Thermocycling with A Stacked Ensemble.

PURPOSE: To investigate the long-term optical properties of CAD/CAM resin-matrix-ceramics (RMCs) under thermocycling and further predict esthetic alteration using machine learning (ML). MATERIALS AND METHODS: Four materials (UP.CAD (UP), Hyramic (Hy), Lava Ultimate (LU), and Vita Enamic (VE)) in high/low translucency (HT/LT) and 0.5/1.0mm (n = 5) were measured at baseline and every 10000 thermocycles up to 40000. L*, a*, and b* were recorded. Whiteness index (WI) and translucency parameter (TP) were calculated and analyzed with t-tests and ANOVA. A stacked ensemble was developed to predict esthetic alteration. RESULTS: Opacities, thicknesses, material types, and thermocycling affected optical properties significantly (P < 0.05). 1.0mm HT VE, 1.0mm LT LU, and 0.5mm HT VE exhibited the highest L*, WI, and TP, respectively. ML predicted esthetic deterioration timings were 18584, 7828, 36577, and 27828 thermocycles for UP, Hy, LU, and VE, respectively, with R2/MSE (L*: 0.9633/0.2375, WI: 0.9724/0.4017, and TP: 0.9801/1.2205). CONCLUSIONS: Compared to UP, RMCs showed greater alterations in L* and WI, lower TP, and ML predicted esthetic deterioration between 7828 and 36577 thermocycles. Notably, LU exhibited superior WI and extended esthetic deterioration timing.