MRI Reconstruction Using Sparse Domain Coefficient Weighing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In magnetic resonance imaging (MRI), reconstructing images from limited data acquired under time-sensitive conditions, such as cardiac motion, is challenging due to insufficient sampling, leading to difficulties in achieving ideal image quality and signal-to-noise ratios with existing compressed sensing techniques.
Innovation Solution
The method involves transforming acquired image data into a sparsity domain, such as a wavelet transform domain, where initial transform coefficients are approximated using compressed sensing techniques like Iterative Hard Thresholding, Orthogonal Matching Pursuit, or reweighted l1 minimization, and then updated based on context information about the proximity of significant coefficients, to construct high-quality images with improved signal-to-noise ratios.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If compressed sensing techniques are used to reconstruct images from limited data, then image reconstruction can be performed with reduced sampling rates, but the image quality and signal-to-noise ratios are insufficient
Solution Approach 1:
The patent applies preliminary action by using an initial estimate of the image (from compressed sensing) to pre-identify locations of significant coefficients before the final reconstruction. This preliminary step guides the subsequent refinement process, allowing the algorithm to focus computational effort on likely significant regions and improve final image quality without requiring more sampling data.
Solution Approach 2:
The patent implements feedback by iteratively refining the image reconstruction using the initial compressed sensing result as feedback. The algorithm uses the initial estimate to identify significant coefficient locations, then uses this information to guide a second reconstruction pass, creating a feedback loop that progressively improves image quality while maintaining the benefits of reduced sampling.
2Reliability
If data acquisition time is shortened to prevent patient motion, then motion artifacts are reduced, but insufficient data is acquired for ideal image reconstruction
Solution Approach 1:
The patent applies parameter changes by modifying the reconstruction process rather than the acquisition parameters. Instead of changing sampling rates or acquisition time, the invention changes the mathematical parameters and algorithms used in reconstruction, enabling high-quality images to be obtained from the limited data that was already acquired during the short time window.
3Measurement precision
If Nyquist rate sampling is used to ensure sufficient data for image reconstruction, then image quality is maintained, but acquisition time increases and patient motion occurs
Solution Approach 1:
The patent applies partial action by using only a portion of the data that would be required for Nyquist-rate sampling. The compressed sensing approach deliberately acquires fewer samples than the traditional Nyquist criterion would require, then uses intelligent reconstruction algorithms to recover the full image quality from this partial data set.
Data Source
AI summary
A method for constructing an image includes acquiring image data in a first domain. The acquired image data is transformed from the first domain into a second domain in which the acquired image data exhibits a high degree of sparsity. An initial set of transform coefficients is approximated for transforming the image data from the second domain into a third domain in which the image may be displayed. The approximated initial set of transform coefficients is updated based on a weighing of where substantial transform coefficients are likely to be located relative to the initial set of transform coefficients. An image is constructed in the third domain based on the updated set of transform coefficients. The constructed image is displayed.