Intraoral Scan Excess Material Removal via ML Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The inclusion of excess material in intraoral scans for dental arches, such as tongue, lips, and tools, occludes clinically significant regions like teeth and gums, requiring manual removal by skilled technicians, which is time-consuming and costly.
Innovation Solution
A method using a trained machine learning model to classify regions in intraoral images, generating probability maps to identify and remove excess material from the images and 3D models, thereby automating the process of material removal.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If manual removal of excess material is performed by lab technicians, then the accuracy of dental arch modeling is improved, but the time and cost of production increase
Solution Approach 1:
The system performs automatic classification and removal of excess material using machine learning models, eliminating the need for manual intervention by lab technicians. The algorithm processes intraoral scans autonomously to identify and remove excess material, achieving both high accuracy and time efficiency simultaneously
Solution Approach 2:
The patent replaces the manual mechanical process of excess material removal with an automated computational system. Machine learning models classify pixels in intraoral scans to automatically identify and remove excess material, substituting human labor with algorithmic processing that maintains precision while dramatically reducing time consumption
2Manufacturing precision
If manual removal of excess material is performed by lab technicians, then the quality of dental scans is improved, but the production cost increases
Solution Approach 1:
The automated system performs excess material removal without requiring skilled lab technician intervention, reducing labor costs while maintaining scan quality. The machine learning model autonomously processes intraoral scans to identify and remove excess material, eliminating the need for expensive manual post-processing
Solution Approach 2:
The patent replaces expensive manual labor with cost-effective automated computational processing. Machine learning algorithms classify and remove excess material automatically, reducing production costs while maintaining or improving scan quality compared to manual methods
3Quantity of substance
If excess material is included in intraoral scans, then the completeness of scan data is improved, but the accuracy of clinical region identification deteriorates
Solution Approach 1:
The system extracts and removes excess material from intraoral scan data while preserving clinically significant regions. The machine learning model identifies excess material pixels and removes them selectively, maintaining the completeness of relevant scan data while eliminating interference with clinical region identification
Solution Approach 2:
The patent applies different classification criteria to different regions of the scan. The machine learning model distinguishes between clinically significant regions requiring high precision and excess material regions to be removed, applying local quality control to maintain accuracy in dental structures while eliminating unwanted elements
4Productivity
If automated machine learning classification is used to remove excess material, then the productivity of scan processing is improved, but the complexity of the system increases
Solution Approach 1:
The patent replaces complex manual processes with automated machine learning systems that, while computationally intensive, streamline the overall workflow. The automated classification and removal process eliminates multiple manual steps, reducing operational complexity despite the sophisticated algorithms employed
Solution Approach 2:
The machine learning system autonomously performs the entire excess material removal process without requiring complex manual intervention protocols. The automated system handles classification, identification, and removal in a single integrated workflow, improving productivity while managing system complexity through self-service automation
Data Source
AI summary
A system includes an intraoral scanner and a computing device. The intraoral scanner generates intraoral scans of a dental site. The computing device processes the intraoral scans using a trained machine learning model to classify points of the intraoral scans into first points having a first dental class and second points having a second dental class; generates a 3D model of the dental site from the intraoral scans; determines first points of the 3D model having the first dental class based on the first points of the intraoral scans and second points of the virtual 3D model having the second dental class based on the second points of the intraoral scans; and removes data for a plurality of the first points of the virtual 3D model having the first dental class to generate a modified virtual 3D model.