MR System Non-Constant Density K-Space Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In contrast agent-intensified 3D MR angiography, achieving high spatial and temporal resolution is challenging due to contradictory requirements, leading to suboptimal image quality and artifacts.
Innovation Solution
A method involving the acquisition of multiple three-dimensional volume data sets in k-space with non-constant density, allowing for flexible reconstruction to optimize temporal and spatial resolution, and image quality by selecting specific data sets post-acquisition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high spatial resolution is achieved by acquiring more k-space data points, then image quality improves, but acquisition time increases and temporal resolution deteriorates
Solution Approach 1:
The patent segments the k-space data acquisition into multiple three-dimensional volume data sets acquired at different time points. Each volume data set contains k-space data points sampled with non-constant density, allowing the system to divide the total data acquisition task into temporal segments. This segmentation enables the system to balance spatial and temporal resolution by selecting which volume data sets to combine based on the specific clinical question being answered.
Solution Approach 2:
The patent applies partial Fourier techniques by acquiring only a subset of k-space data points rather than complete data. Specifically, k-space is sampled with non-constant density, where the degree of sampling varies across different k-space regions. This partial acquisition reduces the total number of data points required while maintaining sufficient image quality through intelligent reconstruction algorithms that can compensate for the missing data.
2Measurement precision
If multiple three-dimensional volume data sets are acquired to improve temporal resolution, then dynamic information improves, but data processing complexity increases
Solution Approach 1:
The patent performs preliminary actions by acquiring multiple three-dimensional volume data sets with k-space data sampled at non-constant density before the final image reconstruction. These volume data sets are prepared in advance with optimized sampling patterns that facilitate subsequent filtering and combination. The preliminary acquisition stage lays the foundation for efficient post-processing by organizing data in a way that simplifies the reconstruction algorithm.
Solution Approach 2:
The patent changes the sampling density parameter across different k-space regions rather than using uniform sampling. By varying the degree of k-space sampling (non-constant density), the system can optimize the balance between acquisition speed and image quality. This parameter change allows fewer total data points to be acquired while maintaining sufficient information for high-quality reconstruction through advanced filtering techniques.
3Productivity
If k-space data is sampled with non-constant density to reduce acquisition time, then acquisition efficiency improves, but image quality may deteriorate
Solution Approach 1:
The patent applies local quality by sampling k-space data with different densities in different regions. Rather than uniform sampling, the system varies the sampling density across k-space, typically acquiring more data points in regions where they provide the most information gain and fewer points where the data is less critical. This localized approach optimizes the trade-off between acquisition efficiency and image quality by concentrating sampling resources where they matter most.
Solution Approach 2:
The patent replaces traditional uniform k-space sampling mechanisms with non-constant density sampling patterns. Instead of systematically acquiring data at regular intervals throughout k-space, the system uses optimized sampling patterns that vary in density. This substitution of the sampling mechanism enables faster acquisition by reducing the total number of data points while maintaining image quality through intelligent reconstruction algorithms that can handle the non-uniform sampling.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables the generation of MR images with improved temporal and spatial resolution, reduced artifacts, and flexible image reconstruction, accommodating various clinical needs.
Implementation Method 1
Magnetic field gradients in chronological sequence and radio-frequency pulses for excitation of the nuclear spins are used for generation of magnetic resonance images
Data Source
AI summary
In a method and apparatus for generation of magnetic resonance images, a number of three-dimensional volume data sets of a subject are acquired in k-space, with each three-dimensional volume data set being acquired with a non-constant density. Filtered three-dimensional volume data sets are generated in k-space, which are assembled from a number of the three-dimensional volume data sets. Three-dimensional image data sets are reconstructed on the basis of filtered three-dimensional volume data sets.