4D Ultrasound Registration via Dominant Motion Vector Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for registering 4D ultrasound volumes often result in inaccurate alignment, leading to poor quality fused images due to inefficiencies in aligning image data sets from different viewpoints.
Innovation Solution
An image data processing method that involves identifying a dominant motion vector in each 4D image data set, rotating and translating the data sets to align these vectors, and then performing a further transformation based on image registration procedures to achieve spatial registration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional optimization processes based on intensity-normalized 2D images are used for registering 4D ultrasound volumes, then the registration process can be performed, but the alignment accuracy is poor and the final fused image quality deteriorates
Solution Approach 1:
The patent applies preliminary action by performing a rough alignment step before the final optimization process. The 4D volumes are pre-aligned using dominant motion vectors derived from reduced resolution data, which provides a better starting point for the subsequent intensity-based optimization, thereby improving final alignment accuracy and fused image quality
Solution Approach 2:
The patent segments the registration process into distinct stages: (1) rough alignment using dominant motion vectors from reduced resolution volumes, and (2) fine optimization using intensity-based metrics. This segmentation allows each stage to address specific aspects of the registration problem, improving overall accuracy
2Loss of information
If multiple motion vectors from different points are used for alignment, then more motion information is captured, but the rotational alignment accuracy decreases due to noise artifacts and extraneous features
Solution Approach 1:
The patent extracts only the dominant motion vector from each 4D volume that represents the most significant motion component, discarding less significant motion information that may correspond to noise or extraneous features. This extraction of the primary motion signal improves rotational alignment accuracy while retaining essential motion information
Solution Approach 2:
The patent changes the parameter of motion representation from multiple vectors to a single dominant motion vector per volume. This parameter simplification, combined with deriving vectors from reduced resolution data, filters out noise while preserving the essential rotational alignment information
3Measurement precision
If full resolution 4D volumes are used for motion vector identification, then the most accurate motion data is obtained, but the processing complexity and computational load increase significantly
Solution Approach 1:
The patent segments the processing by resolution levels, performing motion vector identification on reduced resolution volumes rather than full resolution data. This segmentation reduces computational complexity while the vectors are then applied to align the full resolution volumes, balancing accuracy and processing requirements
Solution Approach 2:
The patent introduces a resolution dimension to the processing strategy by working with multiple resolution levels. Motion vectors are derived from lower resolution versions of the 4D volumes, which reduces the data volume and processing complexity, while the alignment results are then applied to the full resolution data
Data Source
AI summary
An image data processing method is for registering two four-dimensional image data sets, each representative of a time-series of three-dimensional image frames. The method comprises an initial pre-registration step in which 3D image frames of the two image data sets are rotated and translated (16) relative to one another so as to bring into alignment respective dominant motion vectors identified (14) for each, the dominant motion vector being a 3D motion vector representative of motion of an identified three-dimensional sub-region exhibiting maximal spatial displacement over course of the time series.