MRI Fat-Water Separation Using Inversion-Recovery Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Dixon magnetic resonance imaging techniques suffer from phase errors due to B0 inhomogeneities, leading to water/fat swaps and incomplete separation, particularly in areas with high magnetic field inhomogeneity.
Innovation Solution
An additional inversion recovery spin echo scan is performed to identify dominant chemical species (water or fat) per voxel, providing phase information that resolves phase ambiguities and prevents water/fat swaps by using this information in the reconstruction process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If Dixon techniques are used for water-fat separation, then water and fat images can be produced, but phase errors due to B0 inhomogeneity cause water/fat swaps and incomplete separation
Solution Approach 1:
The patent performs a preliminary inversion recovery scan before the main Dixon acquisition to determine T1 values and identify dominant chemical species (water or fat) in each voxel. This preliminary information is then used as a constraint during the Dixon reconstruction process to resolve phase ambiguities and prevent water/fat swaps, thereby improving reliability without sacrificing separation accuracy.
2Reliability
If additional inversion recovery scan is performed to identify dominant chemical species, then water/fat swaps are prevented, but scan time increases
Solution Approach 1:
The patent combines the inversion recovery T1 mapping sequence with the Dixon water-fat separation acquisition by interleaving them. The inversion recovery scan and Dixon sequence share the same hardware resources and are performed in an integrated manner, allowing T1 value determination and dominant species identification to occur concurrently with the main imaging sequence, thereby minimizing additional scan time while maintaining reliable water-fat separation.
3Reliability
If identification map is used as constraint in reconstruction, then water-fat separation robustness is improved, but reconstruction complexity increases
Solution Approach 1:
The patent applies the identification map as a local constraint during Dixon reconstruction by determining the dominant chemical species (water or fat) in each individual voxel based on T1 values from the inversion recovery scan. This per-voxel approach allows the reconstruction algorithm to resolve phase ambiguities locally in each voxel using the pre-determined dominant species information, making the reconstruction process more robust without requiring globally complex modifications to the reconstruction algorithm.
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
The method achieves robust water/fat separation without swaps, even in challenging environments, ensuring accurate imaging results.
Implementation Method 1
The identifying magnetic resonance signals contain an encoding of the dominant chemical species based on differences between longitudinal relaxation times (T1)
Implementation Method 2
phase errors due to B0 inhomogeneities
Implementation Method 3
Dixon methods of magnetic resonance imaging include a family of techniques for producing separate water and lipid (fat) images
Data Source
AI summary
The invention relates to a magnetic resonance imaging system (100). The magnetic resonance imaging system (100) comprises a memory (134) and a processor (130). The memory (134) stores machine executable instructions (140), first pulse sequence commands (142) and second pulse sequence commands (144). Execution of the machine executable instructions (140) by the processor (130) causing the processor (130) to control the magnetic resonance imaging system (100) to acquire identifying magnetic resonance data (146) using the first pulse sequence commands (142). The identifying magnetic resonance data (146) identifies on a per voxel basis, whether the respective voxel is water or fat dominated. Imaging magnetic resonance data (148) is acquired using the second pulse sequence commands (144). A magnetic resonance image (150) is reconstructed using the imaging magnetic resonance data (148). The identifying magnetic resonance data (146) is used to determine on a per voxel basis, whether the imaging magnetic resonance data (148) used for the reconstruction is dominantly induced by water or fat.