K-Space Sampling Pattern Averaging for Low-Ghosting MRI
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
MR imaging techniques face challenges in generating high-quality, high-resolution images with reduced ghosting artifacts and improved signal-to-noise ratio (SNR) using single-shot k-space data acquisition, particularly in cardiac imaging with segmented LGE sequences.
Innovation Solution
A Compressed Sensing (CS) method is employed to acquire multiple single shots of k-space data with different sampling patterns, followed by joint reconstruction of these images to combine and reduce unique artifacts, leveraging inter-image sparsity for improved image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If segmented k-space acquisition is used to achieve high spatial resolution, then image quality and T1-contrast are improved, but ghosting artifacts increase due to patient motion
Solution Approach 1:
The patent applies segmentation by dividing k-space acquisition into multiple single-shot segments, each acquiring a different sampling pattern. This allows the benefits of segmented acquisition (high spatial resolution, good T1-contrast) while mitigating motion artifacts through the use of multiple incoherent sampling patterns that can be combined during reconstruction.
Solution Approach 2:
The patent employs periodic action through repeated single-shot acquisitions with different random sampling patterns. Each shot captures k-space data with a unique incoherent pattern, and these periodic acquisitions are combined through compressed sensing reconstruction to produce high-quality images with reduced artifacts.
2Object-affected harmful factors
If single-shot k-space acquisition is used to eliminate ghosting artifacts, then artifact reduction is achieved, but signal-to-noise ratio and temporal resolution deteriorate
Solution Approach 1:
The patent merges multiple single-shot acquisitions with different incoherent sampling patterns through compressed sensing reconstruction. By combining the information from multiple shots, the system achieves higher signal-to-noise ratio and better temporal resolution while maintaining artifact reduction benefits.
Solution Approach 2:
The patent changes the sampling pattern parameters between shots using different random incoherent distributions. This parameter variation allows the system to eliminate ghosting artifacts while accumulating sufficient signal information across multiple acquisitions to maintain high SNR.
3Object-affected harmful factors
If multiple single shots with different sampling patterns are acquired, then artifact reduction through combining is achieved, but acquisition time increases
Solution Approach 1:
The patent applies partial action by acquiring only the necessary portion of k-space data in each single-shot acquisition using incoherent sampling patterns. Compressed sensing reconstruction then recovers the complete image from these partial measurements, reducing total acquisition time while maintaining image quality and artifact reduction.
Solution Approach 2:
The patent replaces traditional mechanical/sequential k-space filling with a computational approach using compressed sensing reconstruction. This substitution allows multiple single-shot acquisitions to be combined efficiently, reducing artifact content while minimizing the increase in acquisition time through optimized reconstruction algorithms.
Data Source
AI summary
A system and method comprises acquisition of a plurality of k-space sets, each of the plurality of k-space sets comprising a different incoherent variable-density under-sampled combination of k-space data points, performance of iterative reconstruction on the plurality of k-space sets to generate a plurality of images, where each of the plurality of images is associated with a different one of the plurality of k-space sets, and averaging of the generated plurality of images to generate an image.