3D EPI k-space Row Interchange Gradient Pulse Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current 3D echo-planar imaging (EPI) techniques face challenges in achieving shorter recording times while maintaining image quality due to artifacts caused by variable k-space sampling density and gradient strength variations, which lead to spatial frequency-dependent distortions and resolution loss.
Innovation Solution
A method that combines 3D EPI with compressed sensing (CS) by partitioning the k-space into subspaces with equal k-space row distributions and ensuring gradient pulses exceed a threshold value for phase encoding blips, minimizing distortions and artifacts through intelligent gradient pulse variation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If k-space is sampled faster in the read-out direction using EPI, then recording time is reduced, but image distortions arise due to susceptibility differences and gradient field variations
Solution Approach 1:
The patent changes the gradient pulse parameters (strength, timing) dynamically during the echo train to compensate for signal decline and maintain consistent sampling conditions. By adjusting gradient strengths based on the echo position and signal characteristics, the method corrects distortions while maintaining fast EPI acquisition
Solution Approach 2:
The patent implements dynamic gradient pulse adjustment where gradient strengths are varied during the echo train rather than remaining static. This dynamic adaptation allows the system to compensate for T2* signal decay and maintain uniform sampling density across different k-space rows, reducing geometric distortions in the reconstructed image
2Productivity
If variable density Poisson disc sampling is used to achieve compressed sensing, then k-space sampling becomes randomized, but severe image artifacts arise due to small gradient pulses in high density regions
Solution Approach 1:
The patent dynamically adjusts gradient pulse strengths based on the local sampling density requirements. In regions where Poisson disc sampling creates high density (requiring small gradient steps), the system increases gradient strengths to maintain adequate signal-to-noise ratio and avoid artifacts, while in lower density regions it reduces strengths to match the randomized sampling pattern
Solution Approach 2:
The patent applies different gradient pulse characteristics to different regions of k-space based on the local sampling density. By analyzing the Poisson disc sampling pattern and identifying high-density regions, the system applies enhanced gradient strengths locally to those regions while using standard strengths elsewhere, thereby maintaining image quality throughout the field of view
3Adaptability or versatility
If gradient strength is reduced in high density k-space regions to achieve variable density sampling, then compressed sensing is enabled, but resolution loss occurs due to susceptibility-induced gradient fields
Solution Approach 1:
The patent compensates for susceptibility-induced gradient fields by dynamically adjusting the applied gradient strengths. When susceptibility differences create additional gradient fields that would normally require reduced gradient strength, the system increases the gradient strength to counteract these effects, thereby maintaining spatial resolution in the presence of variable density 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
This approach enables significantly shorter recording times with reduced image artifacts, maintaining high image quality by ensuring sufficient gradient strengths and minimizing signal decline between k-space rows, thus overcoming the limitations of previous methods like CAIPIRINHA.
Implementation Method 1
method for recording magnetic resonance data
Implementation Method 2
gradient pulses of the two phase encoding directions for changing between these k-space rows
Data Source
AI summary
Techniques are disclosed for recording magnetic resonance data with a magnetic resonance facility, wherein a three-dimensional echo-planar imaging sequence is used whereby following a single excitation period (e.g. “module”) in an echo train, an echo count of k-space rows is read out in a read-out direction in the k-space, and interchanging takes place between these rows by means of gradient pulses of the two phase encoding directions.