MRI K-space Data Arrangement for Artifact Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Magnetic resonance imaging (MRI) techniques require significant time for data acquisition, especially for time-series data of changing subjects, leading to deteriorated image quality, and existing parallel imaging methods face challenges in efficiently reconstructing high-quality images within shorter times.
Innovation Solution
An MRI apparatus and image processing system that arranges time-series data at specific sampling points in k-space, derives sensitivity distributions in time-space, and generates images using these distributions, allowing for faster imaging by down-sampling and interpolating missing data points, thereby reducing imaging time and improving image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If parallel imaging with down-sampling is used to shorten imaging time, then imaging speed is improved, but image quality deteriorates due to artifacts
Solution Approach 1:
The patent performs preliminary calibration imaging to acquire sensitivity distributions before the actual imaging. These pre-acquired sensitivity maps are then used during the parallel imaging reconstruction process to accurately unfold the down-sampled k-space data, preventing artifacts and maintaining image quality while enabling faster imaging through aggressive down-sampling
Solution Approach 2:
The patent introduces sensitivity distributions (sensitivity maps) as an intermediary element that mediates between the down-sampled k-space data and the final image reconstruction. These sensitivity maps serve as a bridge that enables accurate reconstruction from incomplete sampling data, allowing the system to achieve both fast imaging and high image quality simultaneously
2Manufacturing precision
If calibration imaging is performed to derive sensitivity maps, then image quality is improved, but imaging time increases
Solution Approach 1:
The patent performs calibration imaging only partially - acquiring sensitivity distributions at selected time points rather than continuously throughout the entire time-series imaging. This partial calibration approach provides sufficient sensitivity information for reconstruction while minimizing the time overhead, achieving a balance between image quality and imaging speed
Solution Approach 2:
The calibration imaging is performed preliminarily before the main time-series imaging sequence. By acquiring sensitivity maps in advance at key time points, the system prepares the necessary reconstruction data without interrupting or significantly extending the main imaging acquisition time, thus improving image quality with minimal time penalty
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 significantly shortens the imaging time while maintaining or improving the quality of MR images by deriving sensitivity maps and interpolating missing data, reducing the need for calibration imaging and minimizing artifacts.
Implementation Method 1
A magnetic resonance imaging apparatus (hereinafter, MRI apparatus) is an apparatus that nondestructively visualizes an atom distribution inside a subject body, using a property of atoms such as hydrogen positioned in a magnetic field of selectively absorbing and radiating only electromagnetic waves of a frequency that is dependent on a type of an atom and a magnetic field
Data Source
Figure 1
Figure 2(A)~3(C)
Figure 4
AI summary
A magnetic resonance imaging apparatus (100) according to one embodiment includes an arranger (123a), a sensitivity deriver (123c), and an image generator (123d). The arranger (123a) arranges time-series data at a part of sampling points out of sampling points of a k-space determined based on an imaging parameter. The sensitivity deriver (123c) derives a sensitivity distribution in a time-space, in which the time-series data transformed in a time direction is expressed with a coefficient value, based on the time-series data. The image generator (123d) generates an image of the time-series data using the time-series data and the sensitivity distribution.