Image Registration Using Multi-Level Feature Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The registration of pre- and intra-procedural image data, especially for soft tissue organs, is challenging due to large deformations, leading to underconstrained registration and unwanted deviations from reality.
Innovation Solution
A method for robust and precise image registration involves generating a model dataset based on pre-acquired image data and aligning it with intra-procedural image data using features of different classes at varying levels of detail, incorporating acquisition geometry, and applying transformation rules for rigid and deformable transformations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If registration is based on anatomical and geometrical features commonly mapped in pre- and intra-procedural image data, then the registration can be performed using available data, but the level of detail is limited and large deformations of soft tissue organs lead to large variations in spatial positions causing unwanted deviations from reality
Solution Approach 1:
The registration process is divided into two distinct stages: pre-alignment at a first level of detail using first features, and subsequent registration at a second level of detail using second features. This segmentation allows each stage to operate at appropriate detail levels, improving overall registration precision while managing complexity through structured progression
Solution Approach 2:
A pre-alignment step is performed before the main registration process. This preliminary action establishes an initial alignment at a coarser level of detail, which constrains the subsequent fine registration and prevents large deformations from causing unwanted deviations, thereby improving measurement precision
2Measurement precision
If features at a second level of detail are used for registration, then more precise spatial alignment can be achieved, but the computational complexity and data processing requirements increase
Solution Approach 1:
Features are divided into two classes based on detail level: first features for pre-alignment and second features for precise registration. This segmentation allows the system to process features at different detail levels separately, achieving high spatial alignment precision while managing computational complexity through hierarchical processing
Solution Approach 2:
The pre-alignment at the first level of detail is performed as a preliminary step before registering at the second level of detail. This preliminary action reduces the search space and computational burden for the finer registration, enabling high precision without proportionally increasing overall complexity
Data Source
AI summary
A model dataset is generated based on first image data. The model dataset and second image data map at least a common part of an examination region at a second detail level. The model dataset and the second image data are pre-aligned at a first detail level below the second detail level based on first features that are mapped at the first detail level in the model dataset and the second image data and/or an acquisition geometry of the second image data. The model dataset and the second image data are registered at the second detail level based on second features that are mapped at the second detail level in the model dataset and the second image data. The second class of features is mappable at the second detail level or above. The registered second image data and/or the registered model dataset is provided.