Image Registration Using Anatomical Structures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for registering volume data from x-ray imaging and surface data from optical imaging of a patient's jaw are inaccurate and require mechanical aids or markers, making them impractical for routine use by physicians.
Innovation Solution
A process that aligns tomographically recorded volume data with optically recorded surface data by manually positioning a marking object in both visualizations, using a transformation function to match the extracted volume structure with the surface structure, allowing for automated registration without external references, enabling precise and rapid alignment of optical and x-ray images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If mechanical aids or markers are used for registration, then registration accuracy can be improved, but device complexity and ease of operation deteriorate
Solution Approach 1:
The patent extracts and utilizes inherent anatomical structures (teeth, mandible contours, maxilla boundaries) as natural registration references, eliminating the need for external mechanical markers or aids. This extraction of useful information from the patient's own anatomy enables accurate registration without additional devices.
Solution Approach 2:
The system employs self-service registration by automatically detecting and using the patient's own anatomical features as reference points. The anatomical structures serve themselves as registration markers, eliminating the need for external reference objects and simplifying the overall system.
2Measurement precision
If mechanical aids or markers are used for registration, then registration accuracy can be improved, but ease of operation deteriorates
Solution Approach 1:
The patent extracts and utilizes inherent anatomical structures (teeth, mandible contours, maxilla boundaries) as natural registration references, eliminating the need for external mechanical markers or aids. This extraction of useful information from the patient's own anatomy enables accurate registration without additional devices.
Solution Approach 2:
The system employs self-service registration by automatically detecting and using the patient's own anatomical features as reference points. The anatomical structures serve themselves as registration markers, eliminating the need for external reference objects and simplifying the overall system.
3Ease of operation
If automated registration without markers is implemented, then ease of operation improves, but measurement precision deteriorates
Solution Approach 1:
The patent introduces automated image processing algorithms as intermediaries that bridge the optical and radiological data sets. These algorithms automatically detect anatomical structures, extract feature points, and calculate transformation parameters, enabling precise registration through computational mediation rather than manual marker-based alignment.
Solution Approach 2:
The system replaces mechanical marker-based registration with an automated computational approach. Instead of physically placing and detecting markers, the system uses algorithmic processing of anatomical structures to achieve precise spatial alignment, substituting mechanical operations with digital processing.
4Loss of information
If surface data from optical imaging is combined with volume data from x-ray imaging, then information completeness improves, but device complexity increases
Solution Approach 1:
The patent merges optical surface data and radiological volume data into a unified integrated data set. By combining these complementary data sources through automated registration and fusion algorithms, the system creates a comprehensive representation of the patient's anatomy that includes both surface geometry and internal structural information.
Data Source
AI summary
Disclosed is a method for defining a common reference system in a record of volume data that represents an area of a patient's jaw and is captured by means of an X-ray imaging process and a record of surface data, at least some of which represents the same area of the patient's jaw and which is captured by means of a process for measuring visible surfaces. Volume data and surface data are unhidden on a screen. An object, especially a tooth, which is recognizable in both the volume data and the surface data, is superimposed on each other as congruently as possible in a preliminary positioning step. A volume structure characterizing the object is extracted from the volume data, particularly as a type of edge image, and is made to overlap as much as possible with a corresponding surface structure of the surface data by means of a transformation function, the overlap of the volume structure being adjusted to the surface structure in iterative steps by optimizing a predefined quality level.