MRI Fingerprinting Partial Volume Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Magnetic resonance fingerprinting (MRF) techniques face challenges in accuracy and efficiency due to large dictionaries required for quantitative parameter mapping, which are computationally burdensome and prone to partial-volume effects from mixed tissue types within voxels.
Innovation Solution
Implementing a multi-compartment model for each pixel or voxel and an adaptive reconstruction process to control partial-volume effects and reduce computational burden, using a smaller initial dictionary that coarsely samples acquisition parameters for efficient signal matching.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a large dictionary with fine resolution is used to cover a wide range of tissue parameters, then measurement precision is improved, but device complexity and processing time increase significantly
Solution Approach 1:
The patent divides the large dictionary into multiple smaller sub-dictionaries, each covering a specific range of tissue parameters. This segmentation allows the system to maintain fine resolution within each sub-dictionary while reducing the overall complexity and processing burden compared to a single large dictionary.
Solution Approach 2:
The patent performs preliminary classification of voxels into different tissue types before detailed parameter matching. By pre-categorizing voxels based on initial analysis, the system can select appropriate sub-dictionaries for matching, avoiding the need to search through the entire large dictionary and thus reducing processing time while maintaining accuracy.
2Measurement precision
If a large dictionary is used to ensure correct match for each acquired signal, then measurement precision is improved, but loss of time increases due to significant processing time required
Solution Approach 1:
The dictionary is segmented into sub-dictionaries organized by tissue type and parameter ranges. This allows the matching process to search only relevant sub-dictionaries rather than the entire dictionary, significantly reducing processing time while maintaining matching accuracy through preserved fine resolution in each sub-dictionary.
Solution Approach 2:
The system performs preliminary tissue classification before signal matching. By identifying the tissue type first, the system can pre-select the appropriate sub-dictionary for matching, eliminating the need to search through unrelated tissue parameter ranges and thus reducing processing time without compromising accuracy.
3Device complexity
If voxels are assumed to contain a single tissue type, then device complexity is reduced, but measurement precision deteriorates due to partial-volume effects from mixed tissue types
Solution Approach 1:
The patent segments each voxel into multiple compartments, with each compartment representing a different tissue type. This multi-compartment modeling allows the system to account for mixed tissue types within a voxel, improving measurement precision by accurately representing partial-volume effects while maintaining manageable complexity through the use of sub-dictionaries for each compartment.
Solution Approach 2:
The patent applies different tissue models and parameters to different compartments within a voxel based on their specific tissue characteristics. This local quality approach allows each compartment to be modeled with appropriate precision for its tissue type, improving overall accuracy while keeping the computational complexity manageable through targeted modeling.
Data Source
AI summary
Systems and methods for estimating quantitative parameters of a subject from data acquired using a magnetic resonance imaging (MRI) system. The system includes a computer system configured to control the MRI system to acquire MR fingerprinting (MRF) data representing a plurality of different signal evolutions acquired using different acquisition parameter settings and reconstruct the MRF data into at least one image composed of a plurality of pixels or voxels formed of multiple compartments per pixel or voxel. The computer system is further configured to, on a compartment-by-compartment basis, compare a signal associated with each compartment to an initial dictionary comprising a plurality of signal templates that coarsely sample different acquisition parameters used to acquire the MRF data to determine quantitative parameters for each compartment. The computer system is further configured to generate a quantitative parameter map that indicates the quantitative parameters for each compartment.