Intra-oral Scanner Machine Learning Gingivitis Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for diagnosing gingivitis and periodontal pocket depth are invasive, time-consuming, lack reproducibility, and are not easily adaptable for early detection and monitoring, leading to potential progression of dental diseases.
Innovation Solution
The development of scanner systems and techniques using machine-learning algorithms to analyze intra-oral 3D scan data, providing digital representations of the oral cavity for assessing gingivitis and periodontal pocket depth through modified gingival index and periodontal pocket depth values, enabling non-invasive, efficient, and reproducible detection and monitoring.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional invasive methods are used for diagnosing gingivitis and periodontal pocket depth, then diagnostic accuracy can be achieved, but the procedure becomes time-consuming and lacks reproducibility
Solution Approach 1:
The patent creates a digital 3D copy of the oral cavity using intra-oral scanning technology. This digital model serves as a replica that can be analyzed repeatedly without requiring repeated physical examinations, thereby maintaining diagnostic accuracy while significantly reducing diagnosis time and improving reproducibility
Solution Approach 2:
The patent replaces traditional mechanical probing instruments with optical scanning and machine learning algorithms. The scanner system captures geometric and textural features optically, and AI algorithms automatically analyze these features to diagnose gingivitis and measure periodontal pocket depth, eliminating the time-consuming manual probing process while maintaining or improving measurement precision
2Reliability
If traditional manual assessment methods are used, then clinical evaluation can be performed, but the methods lack reproducibility and are difficult to standardize
Solution Approach 1:
The patent implements an automated system where the scanner and machine learning algorithms perform the assessment independently without requiring manual intervention. The system automatically captures scan data, processes it through AI algorithms, and generates diagnostic results, ensuring consistent and reproducible outcomes while reducing human variability in assessment
Solution Approach 2:
The patent transforms subjective clinical assessments into objective quantitative parameters. By extracting geometric features (volumes, surface areas, depths) and textural features from the digital scan data, the system converts qualitative clinical evaluations into measurable, standardized parameters that improve reproducibility and can be processed automatically
3Reliability
If early detection methods are implemented, then disease progression can be prevented, but the detection systems become more complex and harder to adapt
Solution Approach 1:
The patent creates a multi-functional scanner system that can perform multiple assessments including gingivitis detection, periodontal pocket depth measurement, and monitoring of disease progression. The same digital scan data and machine learning framework are used for different diagnostic purposes, making the system versatile and adaptable to various clinical needs while maintaining early detection capabilities
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
Technologies are disclosed for detection and display of an indication of a gingivitis condition and/or a periodontal pocket assessment of a subject's oral cavity via a digital representation of the oral cavity. Using scan data of the oral cavity, one or more teeth in the oral cavity may be indicated. A first assessment location proximate to a first tooth of the one or more teeth may be indicated. A first image may be generated that may include the first assessment location and one or more first data channels. The one or more first data channels may comprise color data and topological information corresponding to the first assessment location. Using one or more machine-learning algorithms, a modified gingival index (MGI) value and/or a periodontal pocket depth assessment may be determined and displayed for the first assessment location based on the first image and the one or more first data channels.