MRI Phase Correction via K-Space Simulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Magnetic resonance imaging (MRI) systems face challenges in correcting phase variations due to spatio-temporal inhomogeneities of the main magnetic field, leading to image artefacts such as ripples and intensity modulations, which affect image quality despite the use of additional sensors or navigator echoes.
Innovation Solution
A data-driven approach in an MRI system using multiple antenna elements to correct for drifts and fluctuations in the B0-off-resonance field by transforming and combining magnetic resonance data from different channels, allowing for phase correction without additional information, effectively averaging phase errors across neighboring k-space points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If additional sensors or navigator echoes are used to correct phase variations, then correction accuracy is improved, but device complexity and scan time increase
Solution Approach 1:
The patent applies self-service by using the MRI signal data itself to correct phase errors without external sensors. The method reconstructs images from acquired k-space data, simulates expected signals, compares simulated vs. actual signals to detect phase errors, and applies corrections iteratively using only the intrinsic data from the scan, making the system self-correcting without additional hardware
Solution Approach 2:
The patent implements feedback by continuously comparing simulated MRI signals with actually acquired signals, detecting phase errors from this comparison, and applying corrections based on this feedback loop. This closed-loop approach uses the relationship between k-space and image space to generate corrective information that is fed back into the reconstruction process
2Measurement precision
If additional sensors or navigator echoes are used to correct phase variations, then correction accuracy is improved, but scan time increases
Solution Approach 1:
The method uses the existing MRI acquisition data to perform phase correction without requiring additional navigator echoes or separate calibration scans. By leveraging the relationship between k-space and image space transformations, the system extracts phase error information from the data already being collected for image reconstruction
Solution Approach 2:
The patent maintains continuous useful action by performing phase correction as part of the standard image reconstruction pipeline rather than as a separate preprocessing step. The iterative process of reconstruction, simulation, comparison, and correction operates continuously on the acquired data without interrupting or extending the scan
3Measurement precision
If iterative reconstruction with cost-function minimization is used, then correction accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent uses feedback by comparing simulated signals (generated from reconstructed images) with actually acquired signals to detect phase errors. This comparison provides feedback information that guides the correction process without requiring complex cost-function minimization, as the phase errors can be directly extracted from the signal mismatch
Solution Approach 2:
The method creates a simulated copy of the expected MRI signal based on the reconstructed image and compares it with the actual acquired signal. This copying approach allows for straightforward phase error detection by comparing the simulated k-space data with the measured k-space data, avoiding the need for complex iterative optimization
4Device complexity
If phase correction is performed without additional information, then device complexity is reduced, but measurement precision may worsen
Solution Approach 1:
The system performs self-service by using its own acquired data to correct phase errors. The relationship between k-space and image space transformations provides enough information within the MRI data itself to detect and correct phase variations without external references or additional sensors
Solution Approach 2:
The patent implements feedback by continuously comparing simulated MRI signals with actually acquired signals. This internal feedback mechanism allows the system to detect phase errors and apply corrections using only the data already acquired during the scan, maintaining both simplicity and accuracy
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention provides for an MRI system (100) with an RF system for acquiring magnetic resonance data (142). The RF system comprises a set of antenna elements (126). The MRI system (100) further comprises a processor for controlling the MRI system (100). Magnetic resonance data is acquired. Combined image data (144) is reconstructed. The reconstruction comprises transforming the acquired magnetic resonance data (142) from k-space to image space and combining the resulting image data. For each antenna element (126) magnetic resonance data (146) is simulated using the reconstructed combined image data (144). The simulation comprises transforming the reconstructed combined image data (144) from image space to k-space. A phase correction factor is deteremined, The determination comprises calculating phase differences between the acquired magnetic resonance data (142) and the simulated magnetic resonance data (146). The acquired magnetic resonance data (142) is corrected using the phase correction factor. In this way, the invention allows for correcting phase errors caused e.g. by subject motion such as respiration, by B0 off-resonances, etc.