MRI Intensity Inhomogeneity Correction via Convolution Kernel
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Magnetic resonance imaging (MRI) images often suffer from intensity inhomogeneity due to imperfections in the signal acquisition process, leading to smooth intensity variations across the image, which existing technologies have not effectively addressed.
Innovation Solution
A system and method that utilize a processor to generate a convolution kernel from k-space data acquired by both surface and body coils, perform an inverse Fourier transform, and create a corrector to correct intensity inhomogeneity in MRI images by storing the corrector as a data file for electronic adaptation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional MRI signal acquisition is used, then imaging capability is achieved, but intensity inhomogeneity appears in the images
Solution Approach 1:
The patent performs preliminary calibration scans to generate coil sensitivity maps before actual imaging. The convolution kernel is pre-computed from these calibration data, allowing the system to compensate for intensity inhomogeneity in advance rather than attempting to correct it after image acquisition
Solution Approach 2:
The patent introduces a convolution kernel as an intermediary element that mediates between the raw MRI signal and the final image. This kernel, derived from calibration data, acts as a transfer function that corrects intensity inhomogeneity during the image reconstruction process, separating the acquisition imperfection from the final image quality
2Area of stationary object
If multiple coils are used for signal acquisition, then imaging coverage is improved, but intensity inhomogeneity becomes more complex
Solution Approach 1:
The patent merges data from multiple coils by computing a convolution kernel that incorporates sensitivity information from all coils used in the acquisition. This unified kernel is then applied to the combined image data, ensuring that intensity inhomogeneity is corrected consistently across the entire multi-coil imaging coverage area
Solution Approach 2:
The patent generates coil-specific sensitivity maps and convolution kernels for each individual coil, then combines them appropriately. This allows the system to account for the specific intensity characteristics of each coil location, applying localized correction strategies that are then integrated into the global image correction process
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 solution effectively corrects intensity inhomogeneity in MRI images by generating a convolution kernel from k-space data, allowing for improved image quality and accuracy in intensity normalization.
Implementation Method 1
Inverse Fourier transform may be performed on the convolution kernel of the first set of k-space data to obtain an inversely transformed convolution kernel of the first set of k-space data
Data Source
AI summary
The disclosure relates to a system and method for correcting inhomogeneity in an MRI image. The method may include the steps of: acquiring a first set of k-space data, acquiring a second set of k-space data, generating the convolution kernel of the first set of k-space data based on the first set of k-space data and the second set of k-space data, performing inverse Fourier transform on the convolution kernel of the first set of k-space data to obtain an inversely transformed convolution kernel of the first set of k-space data, and generating a corrector based on the inversely transformed convolution kernel of the first set of k-space data. The method may be implemented on a machine including at least one processor and storage.