Joint MRI Slab Reconstruction via Compressed Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional multi-slab MRI acquisitions result in limited incoherence and reduced image quality due to individual slab reconstruction, limiting acceleration potential and image quality in applications like angiography.
Innovation Solution
A method for joint reconstruction of multiple three-dimensional slabs as a single consistent volume using iterative compressed-sensing techniques, applying a cost function with data fidelity terms and regularization through wavelet transforms, allowing for incoherent undersampling and improved signal-to-noise ratio.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multi-slab acquisition is performed with individual slab reconstruction, then acquisition speed is improved through parallel processing, but incoherence is limited and image quality deteriorates
Solution Approach 1:
The patent combines multiple individual slab reconstructions into a single joint reconstruction process. The cost function aggregates data fidelity terms from all slabs simultaneously, and the wavelet transform regularization operates on the entire volume, merging previously separate processing streams to achieve global optimization that improves image quality while maintaining accelerated acquisition.
Solution Approach 2:
The patent transitions from independent 2D slab reconstructions to a unified 3D volume reconstruction. By applying wavelet transforms across the entire 3D volume and enforcing consistency constraints that span all slabs, the method adds a volumetric dimension to the reconstruction process, enabling better exploitation of incoherence and improved image quality.
2Productivity
If accelerated/undersampled acquisition is used, then productivity is improved, but incoherence is limited and image quality may be reduced
Solution Approach 1:
The iterative reconstruction process incorporates feedback loops where the current estimated volume is forward-projected to k-space, compared with actual measured data, and the discrepancy is used to update the estimate. The data fidelity terms enforce consistency between measured undersampled data and the reconstructed volume, with feedback continuing until convergence, thereby maintaining image quality despite accelerated acquisition.
Solution Approach 2:
The patent changes the reconstruction parameters by applying wavelet transforms in the spatial domain and optimizing regularization strength and data fidelity weighting. By adjusting these parameters within the cost function, the method adapts to undersampled data conditions, enabling high acceleration factors while preserving image quality through optimized reconstruction parameters.
3Measurement precision
If thin-slab acquisition is performed to optimize vessel contrast, then angiography quality is improved, but the number of slabs increases and reconstruction complexity increases
Solution Approach 1:
The patent creates a universal reconstruction framework that handles multiple thin slabs simultaneously through a single joint reconstruction process. The cost function and wavelet regularization are designed to work across any number of slabs, providing a multi-functional solution that maintains vessel contrast optimization while managing reconstruction complexity through unified processing rather than separate individual reconstructions.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for acquiring a three-dimensional image volume using a magnetic resonance imaging device includes performing a multi-slice or multi-slab acquisition process to acquire a plurality of slices or three-dimensional slabs corresponding to an imaged object. Each respective slice or three-dimensional slab included in the plurality of slices or three-dimensional slabs comprises k-space data. An iterative compressed-sensing reconstruction process is applied to jointly reconstruct the plurality of three-dimensional slabs as a single consistent volume. The iterative compressed-sensing reconstruction process solves a cost function comprising a summation of individual data fidelity terms corresponding to the plurality of three-dimensional slabs.