K-space Trajectory Synchronization for MR Image Contrast
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing MR image generation methods face challenges in maintaining consistent contrast during data acquisition, leading to artifacts in weighted MR images, particularly when using methods that scan multiple times through k-space, such as radial or segmented spiral methods.
Innovation Solution
The method involves acquiring MR data along uniform trajectories in k-space within a predetermined time period where contrast change proceeds linearly, synchronizing lines before and after a central point in time to ensure consistent contrast, and using pairs of lines with spatial adjacency and temporal symmetry to reconstruct artifact-free images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If SSFP sequence with expanded acquisition window is used for fast MR image generation, then image generation speed is improved, but contrast consistency deteriorates leading to artifacts
Solution Approach 1:
The patent segments the k-space acquisition into multiple uniform trajectories or lines that are acquired in a predetermined order. By dividing the acquisition into discrete, manageable segments with consistent contrast conditions, the method maintains contrast consistency while enabling faster overall image generation through efficient parallel processing of multiple reception coils.
Solution Approach 2:
The patent employs periodic acquisition of k-space lines in a predetermined order, where lines are acquired systematically across multiple trajectories. This periodic structure ensures that contrast changes are uniformly distributed and can be properly weighted during reconstruction, preventing artifacts while maintaining fast imaging through the use of expanded acquisition windows.
2Measurement precision
If radial or segmented spiral methods scanning middle of k-space multiple times are used, then signal-to-noise ratio is improved, but contrast definition deteriorates to undefined mean value
Solution Approach 1:
The patent applies local quality by assigning different weighting factors to different regions of k-space based on their acquisition timing and contrast conditions. Central k-space regions (affecting overall contrast) are weighted differently from peripheral regions (affecting image detail), allowing the method to maintain high signal-to-noise ratio from multiple scans while preserving defined contrast through region-specific optimization.
Solution Approach 2:
The patent dynamically changes the weighting parameters applied to k-space data during reconstruction based on the acquisition time and contrast evolution. By adjusting these parameters, the method transforms the raw data from multiple scans into an image with well-defined contrast that reflects the intended contrast weighting, rather than an undefined mean value.
3Productivity
If contrast changes during data acquisition are allowed, then fast imaging is achieved, but image quality deteriorates due to artifacts
Solution Approach 1:
The patent incorporates feedback mechanisms in the reconstruction process where the known contrast evolution model is used to guide the weighting and combination of k-space data. The reconstruction algorithm continuously adjusts the weighting of acquired lines based on their timing relative to the contrast evolution, ensuring that fast imaging with contrast changes does not compromise image quality but rather produces artifact-free images with the desired contrast characteristics.
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 allows for fast generation of MR images with defined contrast, utilizing all acquired data effectively and reducing artifacts, thereby enhancing the quality of weighted MR images.
Implementation Method 1
Magnetic resonance system, RF excitation pulse, nuclear spins
Implementation Method 2
gradient field system, linear magnetic field gradients in the x-, y- and z-directions
Implementation Method 3
RF reception coils, measurement signals acquired by the RF antenna(s)
Data Source
AI summary
To generate an MR image, acquired MR data are entered into k-space on multiple uniform trajectories in k-space within a predetermined time period. The trajectories are acquired chronologically in a predetermined order before a predetermined point in time, and in a different order after the point in time. The i-th trajectory after the point in time in the different order is adjacent to the (n−i+1)-th trajectory in the predetermined order (n is the number of trajectories acquired before and after the point in time). Two trajectories are adjacent if a distance between them is less than a predetermined threshold. Except for the (n−i+1)-th trajectory, none of the trajectories acquired before the point in time has a distance from the i-th trajectory that is less than the threshold. The predetermined time period is set to be at a middle of a time period after an RF excitation pulse, such that a contrast change within the predetermined time period proceeds as linearly as possible over time.