Digital Beamforming for Parallel MRI Artifact Suppression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing MRI image reconstruction methods fail to effectively combine image pixels from arrays with different types of artifacts, leading to challenges in sensitivity and specificity, particularly in regions with motion and flow artifacts, and do not explicitly account for the directivity of transmit and receive RF elements.
Innovation Solution
The method involves using a beamforming algorithm that accounts for the directivity of RF elements in MRI systems, reconstructing images by determining the directivity of each receive RF element and combining them to produce a final image, which can be implemented in conventional MRI systems without additional hardware, utilizing digital beamforming techniques to preserve information across the field of view and mitigate artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional RSS or adaptive reconstruction methods are used to combine images from multiple RF coils, then image reconstruction is simplified, but the directivity of RF elements is not accounted for leading to artifacts and reduced image quality
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing the directivity patterns of each RF transmit and receive element before image reconstruction. These directivity patterns are computed offline and saved in lookup tables, so that during actual image reconstruction, the pre-computed directivity information can be directly applied without adding real-time computational complexity. This resolves the contradiction by preparing the necessary directional information in advance, enabling accurate artifact suppression while maintaining efficient reconstruction.
Solution Approach 2:
The patent changes the reconstruction approach by incorporating directivity parameters (transmit and receive beam patterns) into the parallel MRI reconstruction process. Instead of using simple summation or adaptive methods that ignore element characteristics, the invention modifies the reconstruction algorithm to weight signals according to pre-computed directivity patterns. This parameter incorporation improves measurement precision by accounting for the actual radiation and reception characteristics of each element, while the pre-computation aspect maintains computational feasibility.
2Reliability
If beamforming algorithms that account for RF element directivity are implemented, then image quality and artifact suppression improve, but computational complexity increases
Solution Approach 1:
The patent resolves the computational complexity issue by performing the computationally intensive directivity pattern calculations in advance, before actual image reconstruction. The transmit directivity patterns (how each transmit element radiates RF energy in different directions) and receive directivity patterns (how each receive element detects signals from different directions) are pre-computed using electromagnetic simulations or measurements, then stored for reuse. This preliminary computation separates the heavy mathematical work from the real-time reconstruction process, maintaining high image quality while enabling practical clinical use.
Solution Approach 2:
The patent uses copying by creating and storing lookup tables of directivity patterns that can be reused across multiple reconstructions. Instead of recalculating directivity patterns for each image or each patient scan, the pre-computed patterns are copied and applied to different datasets. This approach maintains reliable artifact suppression and image quality while dramatically reducing the computational burden during actual reconstruction, as the same directivity information serves multiple reconstruction tasks.
3Object-affected harmful factors
If directivity information is incorporated into the reconstruction process, then artifacts from motion and flow are suppressed, but the reconstruction process becomes more complex
Solution Approach 1:
The patent applies preliminary action by pre-computing the directivity patterns that characterize how each RF element transmits and receives signals in different spatial directions. These pre-computed patterns capture the geometric and electromagnetic characteristics of the coil array, enabling the reconstruction algorithm to properly weight signals from different elements based on their directional sensitivity. This preliminary preparation allows the reconstruction process to suppress motion and flow artifacts effectively while avoiding the need for complex real-time calculations, as the directional weighting factors are already determined.
Data Source
AI summary
Systems and methods for beamforming algorithms for transmit-receive parallel magnetic resonance imaging (“pMRI”) applications are described. For any transmit configuration (e.g., using a single or multiple transmit elements) a weighted sum of the complex image data from each receiver is formed with a spatially-varying weighting. The weighting factor is obtained by solving an optimal refocusing problem at a set of points in the image space, which can include all the pixels in the image. The optimal refocusing of the transmit-receive configuration accounts for the spatially-varying SNR in deriving the coefficients of the weighted sum at every image pixel.