MRI Image Registration via Compressed Candidate Data Sets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image registration methods for MRI scans are inefficient, requiring significant computational resources and time, especially when dealing with moving subjects and image distortions, which hinders real-time registration during image acquisition.
Innovation Solution
The process involves determining a transformation using a compressed data set with a small number of variables, allowing for real-time registration by applying the inverse transformation to the subject data set, significantly reducing computational burden and enabling registration during image acquisition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional image registration methods are used to align MRI images, then registration accuracy is improved, but computational time and resource consumption increase significantly
Solution Approach 1:
The patent segments the large-scale image registration problem into two distinct stages: a coarse alignment stage using a reduced-dimensional representation with fewer control points, and a fine-tuning stage using the full image data. This segmentation allows the computationally intensive accurate registration to be performed only on the refined candidate transformations, dramatically reducing total computational time while maintaining registration accuracy.
Solution Approach 2:
The patent creates multiple low-cost approximate transformations using a reduced-dimensional representation that requires minimal computational resources. These inexpensive candidate transformations serve as disposable intermediates that guide the subsequent accurate registration process, allowing the system to evaluate many potential alignments without the full computational burden of accurate registration for each candidate.
2Manufacturing precision
If traditional image registration methods are used to correct image distortion, then image conformity is improved, but processing speed decreases
Solution Approach 1:
The patent performs preliminary coarse alignment using a reduced-dimensional representation before conducting the final accurate registration. This preliminary action reduces the search space for the accurate registration algorithm, allowing the system to achieve image conformity faster by avoiding exhaustive search through all possible transformations.
Solution Approach 2:
The patent implements a dynamic two-stage registration process that adapts to the specific characteristics of the images being registered. The system dynamically determines the number of control points and the complexity of the reduced-dimensional representation based on the initial assessment of image differences, optimizing processing speed while maintaining conformity accuracy for different registration scenarios.
3Measurement precision
If full-resolution image data is used for registration comparison, then registration accuracy is improved, but computational burden increases
Solution Approach 1:
The patent segments the computational workload by separating the evaluation process into coarse evaluation using reduced-dimensional data and fine evaluation using full-resolution data. This segmentation allows the system to perform initial screenings with minimal computational burden and reserve intensive processing only for the most promising candidate transformations.
Solution Approach 2:
The patent applies partial action by using a reduced-dimensional representation that captures the essential geometric relationships without processing the complete image data. This partial processing provides sufficient information for initial transformation estimation, allowing the system to avoid the excessive computational burden of comparing full-resolution images for every candidate transformation.
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
The present invention relates to a process of bringing at least one subject data set into registration or conformity with a reference data set by electronic methods, each data set being a representation of a respective object. The process comprises: generating each of a plurality of candidate data sets (32) by applying a transformation to a reference data set, the transformation having predetermined variables that are changed such that each of the plurality of candidate data sets is a differently shifted or distorted reference data set; compressing each of the plurality of candidate data sets (34) to form a respective compressed candidate data set and compressing a subject data set (36) to form a compressed subject data set, the step of compressing comprising: determining a plurality of weighting vectors in dependence upon the predetermined variables, the number of weighting vectors being equal to the number of predetermined variables; multiplying all data in a candidate or subject data set by each weighting vector to provide respective, corresponding data elements of the compressed candidate or subject data set; comparing the compressed subject data set with each of the compressed candidate data sets and, in dependence on the comparisons, determining the transformation that has generated the candidate data set corresponding to the compressed candidate data set, which, of the plurality of compressed candidate data sets, provides a best match with the compressed subject data set (38, 40); and applying an inverse of the determined transformation to the subject data set (42).