Dental Caries Detection Using Tooth Edge Reference Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Dentists face difficulties in detecting early signs of dental caries and determining their depth, especially when lesions are confined to the enamel, leading to inadequate non-surgical treatment and reliance on surgical restorative dentistry due to the limitations of current automated analysis methods for dental radiographs.
Innovation Solution
A method that defines areas for analysis using a model fitted to the outer tooth edge, analyzing image variations along elongate profiles parallel to the edge, and combining data from multiple areas to identify indications of dental caries, including the use of Active Shape Models to accurately locate tooth edges and dentinoenamel junctions in radiographs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated analysis methods are applied to process dental radiographs, then productivity is improved, but measurement precision deteriorates
Solution Approach 1:
The patent introduces an intermediary reference model (tooth outline model) that mediates between the automated analysis system and the complex radiograph image. This model serves as a guide to define precise analysis areas, enabling automated processing while maintaining high detection accuracy by focusing computational resources on relevant regions identified by the reference model.
Solution Approach 2:
The patent applies local quality by dividing the radiograph into different analysis areas with different analysis strategies. Areas close to the tooth edge are analyzed with higher precision using multiple reference models, while other areas use standard analysis methods. This localized approach improves overall detection accuracy without sacrificing processing efficiency.
2Measurement precision
If visual examination methods are used to identify carious lesions, then measurement precision is improved for large lesions, but difficulty of detecting and measuring increases for early enamel lesions
Solution Approach 1:
The patent performs preliminary action by first fitting reference models to identify tooth edges and dentinoenamel junctions before analyzing for carious lesions. This preliminary step creates a structured framework that simplifies subsequent lesion detection, reducing the overall complexity of the analysis while improving precision for early enamel lesions.
Solution Approach 2:
The patent segments the analysis into distinct stages: first identifying tooth boundaries using edge detection, then fitting reference models to define analysis areas, and finally analyzing pixel variations within those areas. This segmentation breaks down the complex detection task into manageable steps, reducing overall difficulty while maintaining high precision.
3Device complexity
If analysis areas are defined without reference to tooth structure models, then device complexity is reduced, but measurement precision deteriorates in areas close to tooth edges
Solution Approach 1:
The patent applies partial action by using reference models only in critical areas close to the tooth edge where caries detection is most difficult. In other areas, simpler analysis methods are used. This selective application of complex modeling maintains high precision where needed while minimizing overall system complexity.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
A method for detecting indications of dental caries in an image of a tooth. The method comprises fitting a model to the image to identify a part of the image representing an outside edge of the tooth. Image variations are analysed in an area of the image to detect said indications of dental caries, the area of the image representing a part of the tooth and being defined with reference to the fitted model. Data is then generated indicating the presence of dental caries based upon the analysed image variations.