MRI Artifact Removal via K-Space Peak Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Magnetic resonance imaging (MRI) often produces unwanted image artifacts such as mesh-like or net-like patterns due to local peaks in the measurement dataset, which can obscure real features and reduce image quality.
Innovation Solution
A method that analyzes k space values for local peaks and removes them along with adjacent values based on predetermined threshold criteria, using techniques like Fourier transform and weighting functions to generate a modified dataset that minimizes these artifacts and improves image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional MRI reconstruction methods are used, then image generation is straightforward, but unwanted mesh-like artifacts appear that reduce image quality
Solution Approach 1:
The patent applies preliminary action by analyzing and removing peak values from k-space data before the image reconstruction process. The method identifies local maxima in the measurement dataset and eliminates them prior to Fourier transformation, preventing artifact formation rather than correcting them after reconstruction. This proactive approach improves image quality by eliminating the root cause of mesh-like artifacts before they can manifest in the final image.
2Manufacturing precision
If all k space values are processed for artifact removal, then image quality improves, but computing effort and data volume increase
Solution Approach 1:
The patent applies local quality by focusing computational resources only on specific regions of k-space where peak values are most likely to occur and cause artifacts. Rather than uniformly processing all k-space data, the method identifies and targets local maxima and their surrounding regions for removal. This selective approach maintains image quality improvement while significantly reducing the overall computing effort and data volume that would be required to process the entire dataset.
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
The method effectively reduces unwanted artifacts, enhances image quality, and maintains improved spatial resolution, making it suitable for applications like diffusion or diffusion tensor imaging, while reducing computing effort and data volume.
Implementation Method 1
Imaging by means of magnetic resonance technology, that is to say such as by means of a magnetic resonance tomography unit (MRT)
Implementation Method 2
By means of the data processing facility, the magnetic resonance image is then generated automatically from the modified measurement dataset, that is to say reconstructed or calculated, for example by means of a Fourier transform
Data Source
AI summary
The disclosure relates to a method for generating a magnetic resonance image from a measurement dataset. The measurement dataset is initially acquired from k space values. By means of a data processing facility the k space values are then automatically analyzed at least in terms of their size. Furthermore a modified measurement dataset is automatically generated from the measurement dataset by removing k space values whose size satisfies at least one predetermined threshold value criterion. The magnetic resonance image is then generated automatically from the modified measurement dataset.