K-space Correction for MRI Artifact Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current MRI technologies face challenges in correcting shot-specific phase shifts and magnitude errors in multi-shot diffusion-weighted and SSFP sequence images, particularly when combined with autocalibrated parallel imaging techniques, leading to image artifacts and increased scan time.
Innovation Solution
A method for correcting phase and magnitude errors in multi-shot MRI data acquisition involves acquiring uncorrected k-space data with an autocalibration region and undersampled outer regions, generating corrected k-space data, calculating coefficients based on uncorrected and corrected data points, and synthesizing corrected k-space data using these coefficients to produce artifact-free images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If multiple fully-sampled images are acquired and combined to reduce banding artifacts in SSFP imaging, then image quality is improved, but scan time increases
Solution Approach 1:
The patent segments k-space into different regions (central region and outer regions) and applies different sampling strategies to each. The central region is fully sampled while outer regions are undersampled, allowing parallel imaging acceleration. This segmentation enables artifact reduction in critical regions while maintaining overall scan efficiency.
Solution Approach 2:
The patent applies partial sampling to outer k-space regions while maintaining full sampling in the central region. This partial action approach acquires only the necessary data for artifact correction in the most critical area (center of k-space) while using accelerated undersampling in less critical outer regions, thereby reducing total scan time while still improving image quality.
2Loss of time
If k-space is undersampled with parallel imaging techniques, then scan time is reduced, but image artifacts increase
Solution Approach 1:
The patent applies different quality standards to different regions of k-space. The central region receives full sampling with high quality to ensure accurate phase and magnitude information for artifact correction, while outer regions accept undersampled data with lower quality requirements. This local quality differentiation allows parallel imaging acceleration while maintaining sufficient image quality through targeted correction in the most critical region.
Solution Approach 2:
The patent uses an intermediary calibration region in the center of k-space that is fully sampled to provide reference data for correcting artifacts in the undersampled outer regions. This intermediary region acts as a mediator that enables artifact correction without requiring full sampling of the entire k-space, thereby reducing scan time while maintaining image quality.
3Object-generated harmful factors
If shot-specific phase shifts are corrected for each shot individually before combining, then image artifacts are reduced, but processing complexity increases
Solution Approach 1:
The patent performs phase and magnitude correction on the central k-space region before combining multiple shots. By correcting the most critical region (center of k-space) in advance, the patent eliminates the primary source of artifacts that would otherwise require complex per-shot correction of all k-space regions. This preliminary action on the critical region simplifies overall processing while effectively reducing artifacts.
Data Source
AI summary
A method for performing correction in an autocalibrated, multi-shot MR imaging data acquisition includes performing correction on k-space data in an autocalibration region for each shot individually and then combining the corrected k-space data from each shot to form a corrected reference autocalibration region. Uncorrected source k-space data points are “trained to” the corrected k-space data from the corrected reference autocalibration region to determine coefficients that are used to synthesize corrected k-space data in the outer, undersampled regions of k-space. Similarly, acquired k-space lines in the outer, undersampled regions of k-space may also be replaced by corrected synthesized k-space data. The corrected k-space data from the corrected reference autocalibration region may be combined with the synthesized corrected k-space data for the outer, undersampled regions of k-space to reconstruct corrected images corresponding to each coil element. The corrected images corresponding to each coil element may be combined into a resultant image.