CT Dental Impression Surface Extraction Without Mesh Distortion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for generating digital models from dental impressions are time-consuming, resource-intensive, and prone to inaccuracies due to the use of large triangle meshes and arbitrary iso-value selection, leading to distorted digital surface meshes.
Innovation Solution
A computer-implemented method and system for determining an iso-value of density from a volumetric density file, generating digital surface points, and optimizing the digital surface to create accurate digital models directly from CT scans of dental impressions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional digital surface mesh generation techniques are used, then a digital model can be generated, but the processing time is excessive (up to 10 minutes per model) and computing resources are consumed
Solution Approach 1:
The patent changes the fundamental parameter of how the digital surface mesh is generated by using a point-based approach instead of traditional triangle mesh generation. This involves representing the surface as a collection of points with associated normals and curvatures, which can be processed much faster than generating and then simplifying large triangle meshes. The point-based representation eliminates the need for post-processing simplification while maintaining accuracy.
2Manufacturing precision
If conventional triangle mesh generation is used, then a digital surface mesh can be created, but the mesh becomes distorted and inaccurate requiring post-processing simplification and smoothing
Solution Approach 1:
The patent extracts only the essential geometric information needed for accurate surface representation - specifically points, normals, and curvatures - without generating the full triangle mesh structure. By taking out only the critical elements (points with their geometric properties) and eliminating the complex triangle connectivity, the method achieves high accuracy without the distortions and complexities of conventional mesh generation.
3Ease of operation
If arbitrary iso-value selection is used for digital surface selection, then the process can be simplified, but the surface position, shape, and topology become inaccurate
Solution Approach 1:
The patent implements feedback by using the computed geometric properties (normals and curvatures) of the points to guide the surface selection process. Instead of relying on arbitrary iso-values, the method uses feedback from the actual geometric characteristics of the volumetric data to automatically determine the correct surface position and shape, ensuring topological accuracy without requiring manual iso-value adjustment.
4Productivity
If digital surface mesh simplification and smoothing are applied, then processing can be completed, but a large number of degenerate triangles are created resulting in distorted meshes
Solution Approach 1:
The patent inverts the conventional approach by not starting with a triangle mesh that needs simplification. Instead, it directly generates a point-based representation that inherently avoids the degenerate triangle problem. By working backwards from the final desired outcome (an accurate, simplified surface representation) and using points rather than triangles as the fundamental element, the method eliminates the need for simplification operations that create degenerate triangles.
Data Source
AI summary
A computer-implemented method and system of determining a material surface from a volumetric density file includes generating a density frequency distribution of a volumetric density file of a dental impression and determining an iso-value of density between air and a particular material in the density frequency distribution. A computer-implemented method and system of creating a digital model from a CT scan of a physical dental impression includes selecting an iso-value of density for a digital volumetric density file having one or more voxels, generating one or more digital surface points in virtual 3D space for each of one or more voxels, and selecting a subset of digital surface points from the one or more digital surface points. A computer-implemented method and system of optimizing a digital surface includes moving one or more digital surface points to satisfy a criteria of optimum digital surface selection.


