3D Tooth Segmentation via Interdental Loops for Orthodontic Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Intraoral scanning techniques often generate inaccurate 3D models of teeth due to incomplete light reflection, resulting in image artefacts that impede accurate segmentation and lead to unreliable orthodontic treatment planning.
Innovation Solution
A method involving the identification of interdental loops to segment the crown 3D model from the gingiva and adjacent teeth, using a tooth-gingiva segmentation loop and iteratively determining interdental loops to generate a more accurate 3D representation of the tooth, thereby eliminating image artefacts and improving segmentation accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If intraoral scanning techniques are used to capture 3D models of teeth, then the overall coverage and speed of data acquisition are improved, but image artefacts are generated in interdental spaces due to incomplete light reflection
Solution Approach 1:
The patent applies segmentation by dividing the 3D model into distinct regions using a tooth-gingiva segmentation loop and multiple interdental loops. These loops partition the model to separate individual teeth from gingiva and from each other, eliminating image artefacts in interdental spaces by creating clear boundaries where incomplete light reflection occurred during scanning.
2Device complexity
If traditional segmentation methods are used without interdental loops, then the process is simpler, but the segmented crown 3D model inaccurately represents anatomical features due to image artefacts
Solution Approach 1:
The patent applies preliminary action by generating the tooth-gingiva segmentation loop and multiple interdental loops before performing the actual segmentation of crown 3D models. These pre-established loops define accurate boundaries in advance, ensuring that subsequent segmentation operations produce reliable anatomical representations free from image artefact contamination.
3Loss of time
If crown 3D models with image artefacts are used for orthodontic treatment planning, then the workflow is faster, but the predictability and efficacy of treatment are compromised
Solution Approach 1:
The patent replaces manual image artefact removal or re-scanning with an automated computational method. The processor automatically generates segmentation loops and partitions the 3D model to eliminate artefacts, substituting mechanical re-scanning or manual correction with an algorithmic solution that maintains workflow efficiency while improving treatment planning reliability.
Data Source
AI summary
A method and a system for determining an orthodontic treatment for a plurality of teeth of a subject are provided. The method comprises: receiving a 3D representation of a first tooth and a second tooth, adjacent thereto, of the subject, of a plurality of teeth of the subject; obtaining a tooth-gingiva segmentation loop; identifying an outer set of vertices positioned outside the tooth-gingiva segmentation loop and an inner set of vertices positioned inside the tooth-gingiva segmentation loop; generating a first interdental loop extending, along the 3D representation, through a given inner vertex and a given outer vertex, such that: the first interdental loop is at least partially indicative of an interdental boundary between the first tooth and the second tooth; and the first interdental loop intersects the tooth-gingiva segmentation loop; and generating the first tooth 3D representation based on the tooth-gingiva segmentation loop and the first interdental loop.