RF Shading Correction in 3.0T MRI Using Joint TX/RX Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Higher field MRI systems, such as 3.0 T, face significant challenges in correcting image intensity non-uniformities caused by non-uniform RF fields, leading to shading artifacts that are not adequately addressed by existing methods, which often result in errors exceeding 25% and vary significantly between patients.
Innovation Solution
The proposed solution involves using a geometry model to fit MAP prescan data and decompose it into TX and RX components, generating joint TX/RX correction maps to correct for both transmit and receive RF field non-uniformities without requiring additional hardware or prescans, ensuring anatomical features are preserved and adjustments are made for patient-specific factors like size and tissue composition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If higher field MRI systems (3.0 T) are used to improve image quality and resolution, then image detail and signal strength are enhanced, but RF field non-uniformity increases causing severe shading artifacts and intensity errors exceeding 25%
Solution Approach 1:
The patent performs preliminary mapping scans to characterize RF field non-uniformities before actual imaging. Correction maps are generated in advance by scanning a phantom or reference object, allowing the system to pre-calculate compensation factors that will be applied during image reconstruction to eliminate shading artifacts caused by the non-uniform RF fields at 3.0 T.
Solution Approach 2:
The patent changes the approach from direct imaging to a two-step process: first acquiring correction data under controlled conditions (mapping scan with phantom), then applying mathematical transformations to generate correction factors. These parameter changes allow the system to separate and correct RF non-uniformity effects from actual tissue signal variations, reducing intensity errors from over 25% to under 10%.
2Reliability
If existing correction methods are applied to compensate for RF non-uniformity, then some shading artifacts are reduced, but correction errors remain significant (exceeding 25%) and vary significantly between patients
Solution Approach 1:
The patent applies local quality correction by generating spatially varying correction maps that account for position-dependent RF non-uniformities. Instead of applying a uniform correction factor, the system calculates location-specific correction values based on the mapped RF field distribution, allowing accurate compensation for shading artifacts that vary across different regions of the imaging volume and between different patients.
Solution Approach 2:
The patent implements a feedback mechanism where correction maps are generated based on actual measured RF field distributions from mapping scans. These empirically derived correction factors are then applied to subsequent images, and the system can iteratively refine corrections by comparing corrected images against expected uniformity, continuously improving correction accuracy and consistency across different patients and scanning conditions.
3Measurement precision
If additional hardware or prescans are used to correct RF non-uniformity, then correction accuracy improves, but device complexity and scan time increase
Solution Approach 1:
The patent makes the mapping scan sequence multi-functional by designing it to serve both as a quality control measurement of RF field uniformity and as the source data for generating correction maps. This universal approach allows the same scanning procedure to provide both diagnostic information about system performance and the necessary data for correcting patient images, eliminating the need for separate correction procedures and reducing overall scan time.
Solution Approach 2:
The system performs self-correction by automatically generating correction maps from its own mapping scan data and applying these corrections to patient images without requiring external reference standards or additional calibration hardware. The MRI system uses its内置 capabilities to characterize and correct its own RF non-uniformities, reducing complexity and scan time while maintaining high correction accuracy.
Data Source
AI summary
An MRI MAP prescan data from a predetermined imaged patient volume is decomposed to produce a transmit RF field inhomogeneity map and a receive RF field inhomogeneity map for the imaged patient volume based on a three-dimensional geometrical model of the inhomogeneity maps. At least one of the transmit RF field inhomogeneity map and the receive RF field inhomogeneity map is used to generate intensity-corrected target MRI diagnostic scan image data representing the imaged patient volume.