Point Set Registration Stability on Featureless Surfaces
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The registration of featureless surfaces, such as flat or spherical surfaces, is unstable due to the lack of correspondence in dimensions parallel to the surfaces, leading to registration instability.
Innovation Solution
A method to estimate stability in co-registering datasets by acquiring geometric and additional property data, determining variability, and calculating score factors to predict the success of point-cloud registration using algorithms like ICP and NDT.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If registration is performed on featureless surfaces (flat or spherical surfaces), then the registration procedure becomes unstable due to lack of correspondence in dimensions parallel to the surfaces
Solution Approach 1:
The patent transforms the registration problem from relying solely on 3D geometric correspondence to incorporating 2D image intensity information as an additional dimension. By using the intensity values from pre-operative and intraoperative images, the system creates a multi-dimensional feature space that provides sufficient correspondence even on featureless surfaces, thereby resolving the instability issue.
Solution Approach 2:
The patent introduces image intensity information as an intermediary that mediates the registration process. Instead of directly matching 3D surface points which lack features, the system uses intensity values from images as an intermediate feature representation that enables stable correspondence establishment between pre-operative and intraoperative data.
2Reliability
If additional data such as color or temperature is used to add topology information to surfaces, then registration stability improves, but the method complexity increases
Solution Approach 1:
The patent makes the existing imaging system multi-functional by using it for both anatomical visualization and registration feature extraction. The same images used for surgical planning and navigation also serve as the source of intensity information for stable registration, eliminating the need for separate specialized sensors or complex additional data acquisition systems.
Solution Approach 2:
The system uses its own existing imaging capability to provide the additional intensity information needed for stable registration. Rather than requiring external specialized sensors or complex additional equipment, the registration system leverages the images already captured by the surgical imaging system, making the system self-sufficient.
Data Source
AI summary
The present invention relates to a method encompassing acquiring data as to the geometric structure/topology of a predefined surface section as well as to an additional physical property assigned to the surface section, determining variability of the geometric structure/topology as well as of the additional physical property over the surface section, and determining from these at variabilities an expected stability of co-registering datasets describing the predefined surface section. The present invention further relates to a corresponding computer program and a corresponding system for carrying out this method.