Cartesian K-Space Undersampling for Faster MRI Acquisition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing MRI techniques face challenges in achieving high-resolution imaging with short acquisition times due to the inherent line-by-line scanning scheme and the need for multiple imaging contrasts, leading to prolonged scan times, especially at ultra-high field MRI.
Innovation Solution
A method and system for MRI acquisition using a novel undersampling pattern in the Cartesian k-space, termed ABRICOTINE, which performs undersampled acquisitions in three orthogonal readout directions with random positioning of readout lines, combined with 3D compressed sensing and iterative optimizers for image reconstruction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If line-by-line scanning scheme is used in MRI, then spatial resolution is improved, but acquisition time increases significantly
Solution Approach 1:
The patent applies partial sampling in the phase-encoding direction by acquiring only a subset of k-space lines (e.g., every other line or using random sampling patterns). This partial action reduces the number of measurements needed, directly decreasing acquisition time while using advanced reconstruction algorithms to recover the missing information and maintain spatial resolution.
Solution Approach 2:
The patent changes the sampling parameters by using variable density sampling patterns, random sampling, or non-Cartesian trajectories instead of uniform line-by-line scanning. These parameter changes allow efficient undersampling of k-space while preserving essential spatial information through sophisticated reconstruction methods.
2Loss of information
If multiple imaging contrasts are acquired for clinical exams, then diagnostic information is improved, but total scan time increases
Solution Approach 1:
The patent segments the acquisition process by using parallel imaging with multiple receiver coils, where each coil captures different spatial information simultaneously. This segmentation allows multiple contrasts to be acquired in parallel or with reduced sampling, maintaining comprehensive diagnostic information while reducing total scan time.
Solution Approach 2:
The patent merges multiple contrasts or imaging sequences into a single acquisition using multi-contrast MRI techniques or simultaneous multi-slice imaging. By combining multiple imaging tasks into one scan, the system preserves all necessary diagnostic information while eliminating the time penalty of sequential acquisitions.
3Measurement precision
If higher resolution imaging is performed at ultra-high field, then image quality is improved, but acquisition time increases even more
Solution Approach 1:
The patent introduces parallel imaging across the coil dimension, utilizing the spatial sensitivity profiles of multiple receiver coils as an additional dimension for encoding spatial information. This allows aggressive undersampling in the phase-encoding directions while maintaining high resolution through coil sensitivity information, effectively trading temporal resolution for spatial encoding in a new dimension.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
The invention relates to a method (100) and an MRI system (200) for accelerating MRI acquisitions for imaging a biological object (210), the method comprising: - acquiring (102) MRI data for a region of interest of the biological object; - performing (103) an image reconstruction of said region of interest based on the acquired MRI data; - outputting (104) the reconstructed image of said region of interest; the method being characterized in that it comprises - applying an undersampling pattern of a Cartesian k-space for acquiring said MRI data, wherein undersampled Cartesian acquisitions of MRI data are performed in three orthogonal readout directions, wherein readout lines are positioned along all three spatial orthogonal dimensions of a sampling grid of the Cartesian k-space.