Bone Age Estimation Using Cervical Spine Landmarks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current deep learning-based diagnostic assistance programs in orthodontics are limited to diagnosing existing orthodontic problems and cannot predict the growth stage of patients, lacking the capability to estimate bone age effectively.
Innovation Solution
A bone age estimation method using deep learning models to extract and analyze anatomical landmarks from lateral cephalometric radiographic images, calculating landmark numerical values, and providing maturity information based on these values, including ratios and concavity ratios of the cervical spine, while allowing user input for coordinate changes and visualization of growth progress.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If deep learning models are used to automatically extract landmarks and estimate bone age from lateral cephalometric radiographs, then productivity and ease of operation are improved, but device complexity increases
Solution Approach 1:
The system segments the complex bone age estimation task into distinct modules: a first deep learning model for ROI extraction and a second deep learning model for landmark detection. This segmentation allows each model to specialize in a specific subtask, improving overall efficiency while managing complexity through modular architecture.
Solution Approach 2:
The patent introduces an intermediary processing step where extracted landmarks are used to calculate numerical values (ratios and concavity measurements) before final bone age determination. This intermediary calculation layer simplifies the relationship between raw landmark coordinates and the final diagnostic output, making the system more interpretable and manageable.
2Ease of operation
If automated landmark extraction is implemented using deep learning, then ease of operation is improved, but measurement precision may be compromised without manual verification
Solution Approach 1:
The system implements feedback mechanisms where extracted landmarks and calculated numerical values are presented to clinicians for verification. The maturity information generation provides feedback loops that allow medical professionals to validate automated results, ensuring measurement precision while maintaining ease of operation through automated preprocessing.
3Object-affected harmful factors
If hand skeleton radiographs are replaced with cervical spine analysis from lateral cephalometric radiographs, then harmful factors (radiation exposure) are reduced, but measurement precision may be affected
Solution Approach 1:
The patent makes the lateral cephalometric radiograph multi-functional by extracting not only orthodontic diagnostic information but also bone age estimation data from the same image. This universality eliminates the need for separate hand skeleton radiographs, reducing radiation exposure while maintaining measurement precision through advanced image analysis of the cervical spine region.
Solution Approach 2:
The system transforms the analysis from traditional hand bone metrics to cervical spine parameters (landmark coordinates, ratios, and concavity measurements). By changing the anatomical parameters being measured and the analytical approach, the system achieves accurate bone age estimation from lateral cephalometric radiographs without requiring additional imaging.
4Reliability
If multiple numerical values (ratios and concavity ratios) are calculated from landmarks, then reliability of bone age estimation is improved, but device complexity increases
Solution Approach 1:
The patent enhances reliability by transitioning from single-dimensional landmark coordinates to multi-dimensional analysis including ratios and concavity ratios. This dimensional expansion creates a more comprehensive characterization of cervical spine morphology, improving bone age estimation reliability while the automated calculation system manages the increased computational complexity.
Data Source
AI summary
Disclosed are a bone age estimation method and a bone age estimation apparatus. The bone age estimation method may comprise the steps of: extracting a region of interest including a cervical spine region from a lateral cephalometric radiographic image obtained by imaging a subject's cervical spine, by using a first deep learning model; extracting landmarks from the extracted region of interest by using a second deep learning model; calculating a landmark numerical value on the basis of the extracted landmarks; and providing maturity information of a maturation stage of the cervical spine on the basis of the calculated landmark numerical value.