Dynamic Contrast-Enhanced MRI K-Space Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) techniques face challenges in achieving high spatial and temporal resolution while minimizing noise enhancement and aliasing artifacts.
Innovation Solution
The method involves dividing the k-space into a central and peripheral region, using a 3D k-space trajectory with wave-encoded gradient echoes for controlled aliasing in all three dimensions, and employing a low-rank plus sparse (L+S) reconstruction technique to reconstruct magnetic resonance images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional Cartesian k-space trajectories are used in TWIST imaging, then the acquisition scheme is simple to implement, but the temporal resolution is limited and noise enhancement occurs
Solution Approach 1:
The k-space is segmented into a central region and multiple peripheral regions, which are acquired in an alternating fashion. This segmentation allows for more flexible sampling strategies that can improve temporal resolution while managing complexity through structured organization of the acquisition process.
Solution Approach 2:
The patent transitions from conventional 2D Cartesian k-space trajectories to 3D non-Cartesian trajectories. This dimensional change enables more efficient sampling patterns that can achieve higher temporal resolution by better utilizing the available k-space coverage during the dynamic contrast enhancement process.
2Productivity
If undersampling is performed to increase temporal resolution, then the acquisition speed improves, but aliasing artifacts and noise enhancement occur
Solution Approach 1:
Parallel imaging techniques are introduced as an intermediary method to handle the effects of undersampling. By using multiple receiver coils with different sensitivity profiles, the system can reconstruct images that mitigate aliasing artifacts and reduce noise enhancement, allowing undersampling to be performed without the typical detrimental effects.
Solution Approach 2:
The patent changes the sampling parameters by using non-uniform sampling patterns and varying the density of k-space coverage across different time points. This parameter optimization allows for efficient undersampling that maintains image quality by concentrating sampling efforts where most information is needed while reducing redundant measurements.
3Manufacturing precision
If parallel imaging techniques are applied to control aliasing, then the image quality improves, but the reconstruction complexity and computational load increase
Solution Approach 1:
Coil sensitivity information is acquired and stored in advance during a calibration phase before the actual dynamic imaging. This preliminary action allows the reconstruction algorithm to use pre-computed sensitivity profiles, significantly reducing the computational complexity during the actual imaging and reconstruction process while maintaining the benefits of parallel imaging for aliasing control.
4Manufacturing precision
If 3D k-space coverage is acquired to improve spatial resolution, then the anatomical detail improves, but the acquisition time increases beyond the limited contrast agent evolution window
Solution Approach 1:
The acquisition uses periodic sampling patterns that alternately sample the central and peripheral k-space regions across multiple time points. This periodic action allows the system to accumulate 3D k-space coverage over time while staying within the limited contrast agent evolution window, effectively trading temporal sampling for spatial coverage in a controlled manner.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
Computer-implemented method for determining a dynamic contrast-enhanced magnetic resonance data set of an imaging region of a patient, wherein the k-space (1) to be acquired is divided into a central region (A) comprising the k-space center (4) and a peripheral region (B) surrounding the central region (A), and wherein the peripheral region (B) comprises a partitioning number of peripheral k-space portions (Bi), comprising the steps of - using a plurality of reception coils (18) to acquire magnetic resonance signals using parallel imaging and a k-space acquisition scheme, in which alternatingly the central region (A) and one of the peripheral k-space portions (Bi), the peripheral k-space portion (Bi) being chosen according to a pre-determined order, are imaged in acquisition steps of a pair, such that, after the partition number of such pairs, the whole k-space (1) to be acquired has been imaged and a sliding reconstruction window can be applied to reconstruct an additional magnetic resonance image (10, 11, 12) after each acquisition of such a pair, wherein during acquisition undersampling in at least two k-space directions is performed, - reconstructing a time series of magnetic resonance images (10, 11, 12) forming the magnetic resonance data set from the magnetic resonance signals and sensitivity information regarding the plurality of reception coils (18) by using the sliding reconstruction window and a reconstruction technique for undersampled magnetic resonance data, wherein the k-space trajectories for each acquisition step are chosen to allow controlled aliasing in all three spatial dimensions including the readout direction.