Optical Periodontal Imaging System for Non-Invasive Pocket Depth Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for diagnosing periodontal disease in the oral cavity are invasive, prone to variability based on operator skill, and time-consuming, potentially causing patient discomfort and infection risk, with limited ability to accurately assess disease progression over time.
Innovation Solution
A support apparatus and method utilizing machine learning to derive diagnostic information from three-dimensional image data of the oral cavity, combining optical scanner data and computed tomography data to estimate the positions and relationships of teeth and gingiva, reducing the need for invasive procedures and enabling quicker diagnosis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a hand-held probe is inserted into the periodontal pocket to measure depth, then the depth measurement can be obtained, but the patient experiences pain and bleeding, and there is a risk of pathogenic bacteria entering the blood
Solution Approach 1:
The patent replaces the mechanical probe insertion method with an optical imaging system that uses light to capture images of the periodontal pocket. The imaging device illuminates the pocket and captures reflected light to determine depth, eliminating the need for physical probe insertion that causes pain and bleeding.
Solution Approach 2:
The patent creates an optical copy or image of the periodontal pocket instead of physically measuring it with a probe. By capturing the optical characteristics of the pocket (light reflection, absorption), the system derives depth information from the image data without physical contact that causes harm.
2Measurement precision
If a hand-held probe is inserted into the periodontal pocket, then depth measurement is possible, but pathogenic bacteria may enter the tooth via the probe causing infection
Solution Approach 1:
The patent eliminates the mechanical probe that could transfer bacteria by using optical imaging. The imaging device captures information about the periodontal pocket through light interaction, providing measurement capability without physical contact that could introduce pathogens.
3Measurement precision
If the depth of the periodontal pocket is measured at multiple locations of each tooth, then comprehensive diagnosis is possible, but the time period for measurement becomes long
Solution Approach 1:
The imaging device can continuously capture images of multiple periodontal pockets without interruption. Unlike manual probe measurement that requires insertion and withdrawal at each location, the imaging system maintains continuous operation, capturing all required measurement points in a single or few rapid acquisitions.
Solution Approach 2:
The patent transitions from one-dimensional linear measurement (probe insertion depth) to two-dimensional or three-dimensional imaging that captures multiple measurement locations simultaneously. The image data contains depth information for multiple points across the periodontal pocket landscape, enabling comprehensive assessment in a single capture.
4Measurement precision
If a hand-held probe is used for periodontal examination, then depth measurement can be performed, but the result varies depending on the skills of the operator
Solution Approach 1:
The imaging system performs the measurement automatically through image processing algorithms without requiring operator skill for probe insertion technique. The device captures the image and the processing unit automatically derives depth information, making the measurement process self-sufficient and independent of human operational variability.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A support apparatus (1) includes an input interface (14) that receives input of image data showing a three-dimensional geometry of a biological tissue and a computing device (11) that derives support information including information on positions of at least the tooth and the gingiva relative to each other, by using the image data inputted from the input interface (14) and an estimation model (50) for support of diagnosis of the state of disease in the biological tissue based on the image data, the estimation model being trained by machine learning to derive the support information.