MRI Water-Fat Separation Using Dynamic Field Map Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In dynamic MRI imaging, changes in the static magnetic field non-uniformity map due to factors like respiratory motion and temperature changes lead to phase wrap and local convergence, causing inaccurate water-fat separation and tissue contrast images.
Innovation Solution
An MRI apparatus performs time-series direction discontinuity correction processing in addition to spatial direction correction to address phase wrap and local convergence, ensuring accurate static magnetic field non-uniformity mapping and improved tissue contrast imaging.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If Dixon method is used for water-fat separation in dynamic imaging, then water and fat can be separated, but static magnetic field non-uniformity map changes between time-series images causing phase wrap and local convergence
Solution Approach 1:
The patent performs preliminary phase unwrapping processing on the static magnetic field non-uniformity map before using it for water-fat separation. By preprocessing the map to eliminate phase wrap and local convergence issues, the method ensures accurate and reliable water-fat separation across time-series dynamic imaging without the problems that would otherwise arise from magnetic field non-uniformity changes.
2Measurement precision
If phase unwrapping processing is performed to prevent water and fat replacement, then water-fat separation accuracy improves, but processing complexity increases
Solution Approach 1:
The patent replaces complex iterative mechanical phase unwrapping algorithms with a simplified processing approach that uses the relationship between phase and frequency (phase=2π×frequency×time). By expressing the static magnetic field non-uniformity map as frequency shift and using iterative calculation based on least squares estimation with a signal model, the method achieves phase unwrapping with reduced computational complexity while maintaining accuracy.
3Measurement precision
If iterative calculation based on least squares estimation is used to express static magnetic field non-uniformity as frequency shift, then water-fat separation accuracy improves, but local convergence occurs
Solution Approach 1:
The patent implements feedback mechanisms in the iterative calculation process by using a signal model that includes both water protons and fat protons. The iterative least squares estimation continuously adjusts the frequency shift values based on the difference between measured echo signals and calculated signals, providing feedback that guides the convergence toward the correct solution and prevents local convergence errors.
4Loss of information
If dynamic imaging is performed to monitor contrast agent distribution, then diagnostic information improves, but static magnetic field non-uniformity changes due to respiratory motion and temperature
Solution Approach 1:
The patent addresses magnetic field instability during dynamic imaging by dynamically adjusting parameters in the water-fat separation process. By continuously updating the static magnetic field non-uniformity map using iterative least squares estimation with frequency shift expression, and applying phase unwrapping processing to each time point, the method maintains accurate water-fat separation despite changes in magnetic field conditions caused by respiratory motion and temperature variations during contrast agent monitoring.
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 enhances the accuracy of tissue contrast images by reducing phase wrap and local convergence, leading to more precise water-fat separation and stable static magnetic field non-uniformity mapping, even in dynamic imaging scenarios.
Implementation Method 1
The Dixon method is an imaging method using a phase rotation difference between water protons and fat protons at each TE for acquiring an echo signal
Implementation Method 2
using a difference in the behavior of water protons and fat protons after the application of a high frequency magnetic field
Implementation Method 3
a gradient magnetic field power source 107, a bed 106 on which an object 101 is placed, an irradiation coil 104 for irradiating the object 101 with a high frequency magnetic field pulse
Data Source
AI summary
To acquire a more accurate tissue contrast image such as a water-fat separation image in time-series imaging by an MRI apparatus, a static magnetic field non-uniformity map is created from a plurality of echo signals having different TEs obtained in time series, and a water-fat separation image is obtained using the static magnetic field non-uniformity map. Processing (spatial direction discontinuity correction processing) for correcting discontinuity of a phase or a frequency in a spatial direction and processing (time-series direction discontinuity correction processing) for correcting discontinuity in a time-series direction are performed for the static magnetic field non-uniformity map.