Key Technologies Research For Modeling Inlay In Dental Restoration

In recent years, CAD/CAM has gained wide application in the area of dental restoration. Compared with the traditional hand-made method, dental CAD/CAM technology not only improves the quality of the restoration tremendously, but also shorten the production period significantly, meanwhile, it also meet the people’s individual requirements. Several key algorithms have been designed and a whole framework has been proposed in this dissertation which mainly aims at the surface modeling of the inlay restoration.The theories of computer aided geometric design, discrete differential geometry, computer graphics, computational geometry, probability theory and reverse engineering are all incorporated into our algorithms. Main contents and contributions of the dissertation include:1. Traditional region growing algorithms have some limitations when the point clouds are non-regular and when the data sets are very large, these surface reconstruction methods are not resilient to non-uniform point cloud and the processing speed of most of them is not fast enough.To overcome these limitations, a novel surface reconstruction algorithm based on Hash-Grid spatial subdivision scheme and adaptive region growing strategy is proposed in this dissertation. The algorithm consists of the following two steps:in order to increase the triangulation efficiency, a novel data structure named Hash-Grid for computing the k-nearest neighbors is proposed; in order to improve the quality of the triangle mesh, the searching radius can be determined adaptively according to the point uniformity during mesh growing, at the same time,a topological examination criterion is also defined ensuring the candidate triangle would not be non-maniflod. The algorithm can reconstruct a topological correct surface from unorganized points in a fast and reliable manner.2. The extraction of inlay margin line is one of a key link during inlay modeling, existing global feature extraction algorithms have some shortcomings to detect the margin line, so a novel inlay margin line extracting algorithm based on heuristic searching strategy is proposed in this dissertation. Combing with analyzing the feature information of the triangle mesh based on local surface fitting method and exploring the shortest path problem in graph theory, an initial margin line is extracted by comparing the value of the heuristic cost function which is defined according to the featue information, then the margin line is optimized to ensure its smoothness and closed quality.3. A statistical crown reconstruction method for restoring the missing surface of a tooth crown is presented in this dissertation. A mathematical representation of crown morphology can be defined by a quantitative formulation and the missing surface regeion can be infered with a3D statistical deformable model which is determined by the crown model parameters and its corresponding eigenvectors. In contrast with the standard crown database which can only represent limited crown morphology, the statistical crown describe various anatomical morphology so that the pose and anatomical features of the tooth will work well for chewing without need more complicated occlusion inspection.4. For modeling the occlusion surface of inlay, a multi-weighting constrained Laplacian deformation algorithm is proposed in this dissertation. The damaged tooth surface can fit well with the statistical crown after adaptive deformation. Two additional deformation algorithms based on boundary constraint and minimal thickness constraint are put forward to achieve a good marginal discrepancy and deformation result.