Multi-Resolution QSM MRI Artifact Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current quantitative susceptibility mapping (QSM) techniques using MRI are plagued by inaccurate results and spatial 'streaking' artifacts due to ill-conditioned inversion algorithms, requiring substantial computational time and being sensitive to heterogeneities and patient motion, limiting their clinical utility.
Innovation Solution
A method involving multiresolution processing of magnetic resonance data using parcellated field shift maps, where the data is divided into subvolumes and processed using machine learning algorithms to generate quantitative susceptibility maps, reducing artifacts and improving computational efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If iterative QSM algorithms are used to invert the forward model, then the quantitative susceptibility map can be obtained, but substantial computational time is required and streaking artifacts occur
Solution Approach 1:
The patent divides the field shift map into multiple resolution layers (coarse to fine) and processes each layer separately. This segmentation allows the inversion algorithm to work on smaller, more manageable subsets of data at each resolution level, reducing overall computational time while maintaining accuracy through progressive refinement from coarse to fine details
Solution Approach 2:
The patent performs preprocessing steps including background field removal and phase unwrapping before the main inversion process. By preparing the data in advance and removing known artifacts beforehand, the main inversion algorithm operates on cleaner data, reducing computational burden and improving efficiency without sacrificing accuracy
2Reliability
If conventional QSM inversion algorithms are applied to clinical datasets, then susceptibility maps are generated, but large signal fluctuations and poor data consistency occur due to heterogeneities and patient motion
Solution Approach 1:
The patent applies different processing strategies to different regions of the field shift map based on local characteristics. By dividing the map into resolution layers and processing local regions separately, the algorithm can adapt to local heterogeneities and motion artifacts, improving data consistency in each region while reducing streaking artifacts through localized inversion
Solution Approach 2:
The patent introduces a resolution layer dimension to the processing, transforming the problem from a single-resolution inversion to a multi-resolution hierarchical inversion. This additional dimensional approach allows the algorithm to capture both global and local features, improving robustness against heterogeneities and motion while reducing artifacts
3Measurement precision
If manual processing and modifications are applied to reduce artifacts, then data quality improves, but the process is not automated and requires substantial time
Solution Approach 1:
The patent implements an automated multi-resolution inversion algorithm that performs preprocessing, inversion, and artifact reduction without manual intervention. The system automatically divides the field shift map into resolution layers, applies the inversion algorithm at each level, and combines results, eliminating the need for manual processing while maintaining high data quality and consistency
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 constrains residual artifacts to individual volumes, enhances computational efficiency, and enables real-time QSM, providing more reliable and accurate susceptibility maps with reduced computational time.
Implementation Method 1
Magnetic resonance data acquired with an MRI system are provided to a computer system, and a field shift map is generated from the magnetic resonance data
Implementation Method 2
These algorithms seek to invert a known 'forward model' dipole convolution that relates the measured axial component of the MRI magnetic field with an isotropic magnetic susceptibility tensor
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
Systems and methods for quantitative susceptibility mapping ("QSM") using magnetic resonance imaging ("MRI") are described. Localized magnetic field information is used when performing the inversion to compute quantitative susceptibility maps. The localized magnetic field information can include multi-resolution subvolumes obtained by segmenting, or dividing, a field shift map. In some instances, a trained machine learning algorithm, such as a trained neural network, can be implemented to convert the localized magnetic field information into quantitative susceptibility data. These local susceptibility maps can be combined to form a composite quantitative susceptibility map of the imaging volume.