MRI Chemical Shift Artifact Correction via Adipose Voxel Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Magnetic resonance imaging (MRI) systems face challenges with intensity inhomogeneity and chemical shift artifacts, particularly in phase-sensitive reconstruction methods like the Dixon method, which affect image quality and segmentation accuracy, especially in thin structures and fat-water separation.
Innovation Solution
An iterative method is developed to correct for chemical shift artifacts and intensity inhomogeneity by generating an intensity correction field using pure adipose tissue voxels, allowing for accurate fat content estimation and image stabilization, even after contrast agent injection, and enabling quantitative measurement of different types of fat without partial volume effects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If phase sensitive acquisition and reconstruction (Dixon method) is used to achieve effective fat and water separation, then fat and water separation accuracy is improved, but chemical shift artifact (water-fat shift artifact) occurs causing spatial misregistration
Solution Approach 1:
The patent introduces an intermediary intensity correction field that mediates between the fat signal and the final water image. This correction field, derived from pure fat voxel identification and intensity analysis, compensates for the spatial misregistration effects of chemical shift artifacts while preserving the accurate fat-water separation achieved by the Dixon method.
Solution Approach 2:
The patent performs preliminary identification of pure fat voxels and generation of the intensity correction field before final image reconstruction. By pre-characterizing the intensity inhomogeneity patterns in fat regions and creating the correction field in advance, the system can apply compensatory adjustments during water image generation without compromising the fundamental fat-water separation accuracy.
2Stability of the object's composition
If intensity correction field is generated using pure fat voxels to correct intensity inhomogeneity, then image intensity uniformity is improved, but operator time for manual voxel identification is increased
Solution Approach 1:
The patent implements self-service by enabling the system to automatically identify pure fat voxels and generate the intensity correction field without operator intervention. The method uses algorithmic criteria (signal intensity thresholds, spatial distribution patterns) to autonomously characterize fat regions and compute correction parameters, eliminating manual voxel identification while maintaining correction effectiveness.
Solution Approach 2:
The patent transforms the correction approach by changing from manual parameter specification to automated parameter derivation. By adjusting the identification criteria to use signal intensity parameters and spatial distribution characteristics that can be automatically computed, the system converts an operator-dependent process into an autonomous computational task that maintains intensity uniformity correction.
3Manufacturing precision
If flyback protocol is used to eliminate chemical shift artifact, then spatial misregistration is reduced, but signal to noise ratio (SNR) decreases
Solution Approach 1:
The patent extracts the harmful chemical shift artifact effects from the imaging process by separately identifying and correcting them through the intensity correction field. Instead of using flyback protocol that re-acquires data to eliminate artifacts, this method extracts the artifact characteristics from the initial acquisition and removes them computationally, preserving the original high-SNR signal data.
Solution Approach 2:
The patent creates a computational copy of the intensity correction information from pure fat voxels and applies it to correct the water image. This copying approach allows the system to simulate the effect of artifact elimination without re-acquiring data, thereby maintaining the original signal-to-noise ratio while achieving spatial accuracy improvement.
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 effectively reduces chemical shift artifacts, improves image quality by stabilizing signal intensity, and allows for precise fat content measurement in each voxel, reducing operator dependence and partial volume effects, thus enhancing the accuracy of fat quantification in MRI images.
Implementation Method 1
The resonance frequency of protons in human methylene lipid [CH2]n and water differ by 3,5 ppm corresponding to 224 Hz at a field strength of 1.5 T. This intrinsic difference can be utilized for effective fat and water separation using the phase sensitive acquisition and reconstruction.
Implementation Method 2
Phase sensitive acquisition and reconstruction, such as an 'in-phase sand out-of-phase'-method of which the Dixon method may be best known
Implementation Method 3
In magnetic resonance imaging (MRI) the frequency is used to encode the spatial position of the signal. As the RF-pulse is tuned at the frequency of water, fat will have a relative frequency shift that cannot be distinguished from the phase difference introduced by the frequency encoding.
Implementation Method 4
a consequence from utilizing resonance frequency shift in such methods is spatial misregistration in the frequency encoding direction known as the chemical shift artifact, in the case of fat and water images, the water-fat shift artifact
Data Source
Figure 1a~1f
Figure 2
Figure 3
AI summary
Present invention discloses systems and methods for improvement of magnetic resonance images. Correction of a chemical shift artefact in an image acquired from a magnetic resonance imaging system is obtained by a system and a method involving iterative - compensation for the misregistration effect in an image domain. Correction of an intensity inhomogeneity in such images is obtained by a system and a method involving locating voxels corresponding to pure adipose tissue and estimating correction field from these points.