3D Dentition Model Refinement for Interproximal Space Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional 3D models of a patient's dentition generated by optical scanners often suffer from inaccuracies in resolving interproximal spaces due to incomplete or erroneous scan data, leading to errors in orthodontic treatment planning.
Innovation Solution
Methods and apparatuses refine the 3D model by using 2D images to identify interproximal spaces and generate new surface points, smoothing edges, and removing points within these spaces, utilizing techniques like structured light imaging and neural networks to enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional optical scanners are used to generate 3D models of dentition, then the scanning process is simple and fast, but the accuracy of interproximal space representation deteriorates due to incomplete or erroneous scan data
Solution Approach 1:
The patent combines multiple data sources (intraoral scan data, photographs, and video recordings) to create a comprehensive 3D model. By merging these different types of data, the system overcomes the limitations of individual scanning methods and achieves accurate representation of interproximal spaces while managing complexity through integrated processing
Solution Approach 2:
The patent uses machine learning algorithms and automated processing systems as intermediaries to bridge the gap between raw scan data and accurate 3D models. These intermediary systems automatically identify and correct errors in scan data, particularly in difficult-to-scan interproximal regions, improving accuracy without requiring manual intervention
2Measurement precision
If intraoral scanners are used to capture dentition data, then the scanning process is efficient, but the resolution of interproximal regions deteriorates due to shadowing and obscuration from external imaging angles
Solution Approach 1:
The patent performs preliminary processing of scan data to identify regions that are difficult to scan, such as interproximal spaces. By detecting these problematic areas early in the process, the system can apply targeted corrections and additional imaging techniques specifically to these regions, improving resolution without requiring extensive re-scanning of the entire dentition
Solution Approach 2:
The patent transitions from 2D photographs and video recordings to 3D modeling by projecting 2D image data onto 3D surfaces. This dimensional transformation allows the system to recover information from shadowed interproximal regions by using multiple 2D views and reconstructing the 3D geometry, thereby improving resolution without additional scanning time
3Reliability
If 3D capture methods are used to define tooth surfaces, then the process is straightforward, but the reliability of interproximal space detection deteriorates with resolution limits of more than 300 μm
Solution Approach 1:
The patent implements feedback mechanisms where the system continuously evaluates the quality of scan data and automatically identifies regions that require improved definition. By using machine learning algorithms to assess detection reliability, the system can trigger additional processing steps or request re-scanning of specific areas, thereby improving both detection reliability and manufacturing precision through iterative refinement
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The refined 3D models provide more accurate representations of interproximal regions, enabling better orthodontic treatment planning with improved resolution and clarity of interproximal spaces.
Implementation Method 1
obtain a plurality of individual three-dimensional surface data sets of surfaces of the patient's teeth using an intraoral scanner
Implementation Method 2
the three-dimensional model may be determined with a non-contact, optical scanner
Data Source
AI summary
An apparatus (system, device, method, and the like) is disclosed for refining a three-dimensional (3D) model, particularly 3D models of a subject's dentition. An initial 3D model is received or generated along with a plurality of two-dimensional (2D) images corresponding to the 3D model. The 3D model is refined using edge boundaries of a space around or between two or more objects of the 3D model identified from the 2D images.