Hybrid 3D Teeth Segmentation via Iterative Mesh Refinement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for segmenting 3D mesh images of teeth from gums and other supporting structures are often inaccurate and require significant operator skill, computational complexity, and multiple radiation-exposing scans, failing to distinguish tooth structures effectively, especially in cases of varying tooth shapes and positions.
Innovation Solution
A computer-implemented method for generating segmented 3D teeth models using a hybrid approach that combines automated segmentation with operator interaction, allowing for iterative refinement and correction of segmentation results, utilizing patterned light imaging and polygon mesh data to differentiate teeth from gums and other structures without requiring multiple CBCT scans.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional segmentation methods are used on 3D mesh images, then segmentation can be performed, but the accuracy of tooth identification is poor and operator skill is required
Solution Approach 1:
The system performs automated segmentation where the computer processor automatically identifies and segments teeth from the 3D mesh model without requiring operator intervention or skill. The algorithm independently processes the mesh data to generate segmented tooth models, eliminating the need for skilled operators to perform manual segmentation while maintaining high accuracy through iterative refinement processes.
2Measurement precision
If multiple CBCT scans are used to improve segmentation accuracy, then tooth structure differentiation improves, but patient radiation exposure increases
Solution Approach 1:
The system uses optical surface contour imaging to create a digital copy or model of the tooth surfaces, which then serves as the basis for 3D mesh generation and segmentation. This optical copying approach replaces the need for multiple radiation-exposing CBCT scans, as the segmented 3D mesh model derived from optical data provides sufficient detail for accurate tooth identification without additional radiation exposure to the patient.
3Productivity
If automated segmentation is used to reduce operator time, then processing speed improves, but segmentation accuracy decreases
Solution Approach 1:
The system employs iterative segmentation procedures that continuously refine the segmentation results. The computer processor executes multiple segmentation iterations on the 3D mesh model, with each iteration improving the accuracy of tooth identification. This continuous automated processing maintains high productivity while achieving high segmentation accuracy through successive refinements without requiring operator intervention between iterations.
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
This method provides accurate and efficient tooth segmentation with reduced operator time and computational complexity, improving the accuracy of tooth identification and reducing patient radiation exposure by leveraging both automated processing and human perception.
Implementation Method 1
Fringe projection imaging uses patterned or structured light and camera/sensor triangulation to obtain surface contour information for structures of various types.
Data Source
AI summary
A computer-implemented method for generating one or more segmented 3-D teeth models obtains a 3-D mesh model of a patient's dentition and executes a first segmentation procedure on the obtained 3-D mesh model, displaying one or more segmented teeth from the 3-D mesh model. At least one of the one or more segmented teeth is recorded according to operator instruction and removed from the 3-D mesh model to form a modified 3-D mesh model. A repeating sequence identifies a modified segmentation procedure, executes the modified segmentation procedure on the modified 3-D mesh model, displays one or more segmented teeth from the modified 3-D mesh model, records at least one of the one or more segmented teeth, and removes the recorded at least one tooth from the modified 3-D mesh model. Recorded segmentation results are displayed, stored, or transmitted.