Modelos digitais : comparação do escaneamento em diferentes angulações e de 4 métodos de sobreposição

Abstract

Study models are an essential part of Orthodontic diagnosis and treatment planning. Traditionally, diagnostic measurements have been obtained from plaster models, but the technological advancements have made the introduction of digital models into dental practices. Digital models have the advantage of allowing a “virtual setup”, and the superimposition of different stages of Orthodontic treatment. Objective: To review the literature on the present stage of digital models in Orthodontics, and to assess the reliability and reproducibility of linear measurements in digital models, obtained at different angulations, and of 4 methods of superimposition using reference points and areas. Methods: 28 maxillary plaster models were selected and scanned with R700 scanner (3Shape, Copenhagen, Denmark), in four different angulations (0º, 5º, 10º and 15º in relation to the occlusal plane) yielding 4 groups (0, 5, 10 and 15). All analysis and measurements were done with OrthoAnalyzer Software (3Shape). The digital models, in stl file format (stereolithography), were used to perform the linear measurements, assessing the central and lateral portions of the maxillary model. Additionally, for reproducibility evaluation, 4 methods of superimpositions were tested: 1 (3 central points), 2 (3 lateral points), 3 (surface and 1 point) and 4 (surface and 3 points). Data were evaluated using intraclass correlation coefficients (ICC), ANOVA (1-way) and Kappa coefficient. Results: All 4 groups showed good correlation with each other, with no significant difference among them, for all linear measurements. Evaluated the best superimposition method correlation was observed between methods 3 – 4, indicating that to superimpose with accuracy, is necessary to select points and a surface. The first superimposition method, that used 3 central points, showed slight correlation with the other methods, being unrealible to superimposition purpose. Conclusions: There were no statistical difference between linear measurements of digital models at different scanning angulations. Regarding the superimposition methods, the sufarce based ones (methods 3 and 4) were the best.