Motion-Corrected Parameter Mapping for Cardiac MRI
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current magnetic resonance imaging techniques, such as Model-Based Acceleration of Parameter Mapping (MAP), are not suitable for cardiac imaging due to anatomical motion, which prevents accurate selection of inversion time (TI) for late gadolinium enhancement imaging, impeding differentiation between healthy and diseased tissue.
Innovation Solution
The integration of a forward-backward motion correction (MOCO) method with the MAP procedure allows for the reconstruction of fully sampled 2D images, enabling the determination of T1* maps and retrospective selection of optimized image contrast, even in the presence of cardiac motion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If Model-Based Acceleration of Parameter Mapping (MAP) is used for cardiac imaging, then parameter mapping capability is improved, but anatomical motion causes reconstruction inaccuracies and prevents accurate TI selection
Solution Approach 1:
The patent applies preliminary motion correction by estimating anatomical motion from fully sampled images before performing parameter mapping reconstruction. This preliminary action removes motion-induced inconsistencies from the undersampled data, allowing the MAP technique to achieve accurate parameter maps despite cardiac motion during acquisition
Solution Approach 2:
The patent introduces motion estimation and motion correction as intermediary steps between data acquisition and parameter mapping reconstruction. The motion estimation acts as a mediator that characterizes the anatomical motion, and the motion correction uses this characterization to compensate for motion effects in the reconstruction process
2Productivity
If undersampled radial single-shot measurements are used for acceleration, then acquisition time is reduced, but undersampling artifacts and motion-induced inconsistencies deteriorate image quality
Solution Approach 1:
The patent performs preliminary motion estimation using the fully sampled images acquired during the accelerated radial sampling process. This motion estimation is then applied to correct motion-induced inconsistencies in the undersampled data during reconstruction, enabling fast acquisition without sacrificing image quality or parameter mapping accuracy
Solution Approach 2:
The patent changes the sampling strategy from uniform undersampling to motion-aware reconstruction. By incorporating motion estimation parameters into the reconstruction process, the system can reconstruct accurate parameter maps from undersampled data that would otherwise be corrupted by motion artifacts
3Ease of operation
If manual TI selection is used for LGE imaging, then workflow simplicity is maintained, but tissue contrast differentiation is impeded due to sub-optimal TI selection
Solution Approach 1:
The patent implements feedback by using the reconstructed T1* parameter maps to inform and optimize the TI selection for LGE imaging. The system reconstructs parameter maps first, then uses the derived T1* values to calculate and select the optimal TI that maximizes tissue contrast, creating a feedback loop where reconstruction results improve subsequent imaging parameters
Solution Approach 2:
The patent performs preliminary parameter mapping and T1* calculation before the actual LGE image acquisition. This preliminary action provides the information needed to select the optimal TI, transforming the workflow from manual guesswork to data-driven optimization while maintaining overall process efficiency
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
This approach enables the use of parameter maps to select optimal TI for LGE imaging, improving tissue contrast differentiation and image quality by stabilizing the reconstruction process and removing undersampling artifacts, thus facilitating the evaluation of myocardial infarction.
Implementation Method 1
Magnetic Resonance Imaging (MRI), also called magnetic resonance tomography, is a known imaging modality wherein an object to be examined, such as a patient in the case of medical MRI, is placed in a magnetic resonance data acquisition scanner
Implementation Method 2
As the excited nuclear spins relax, they emit radio-frequency signals, called magnetic resonance signals
Implementation Method 3
Gradient magnetic fields are superimposed on the basic magnetic field that respectively change in strength along the three axes of a Cartesian coordinate system, so that the received magnetic resonance signals are spatially encoded
Implementation Method 4
The values in k-space can be mathematically transformed, such as by the use of Fourier transformation, into values of pixels in image space, in a procedure known as image reconstruction
Implementation Method 5
The integration of a forward-backward motion correction (MOCO) method with the MAP procedure allows for the reconstruction of fully sampled 2D images, enabling the determination of T1* maps
Data Source
AI summary
In a magnetic resonance (MR) method and apparatus, an image reconstruction algorithm is used to reconstruct image data from k-space data that represent an acquired MR signal, and the reconstruction algorithm makes use of a model that requires the MR signal to exhibit a signal behavior from a relaxation model. In order to permit the reconstruction algorithm to be used when the acquired MR signal does not exhibit the model signal behavior due to motion of the subject, the k-space data are motion-corrected so as to produce corrected k-space data that represent said model signal behavior, and image data are reconstructed from the corrected k-space data using the reconstruction algorithm.