MRI Quantitative Parameter Mapping via Low-Rank MT-Aware Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current magnetic resonance imaging (MRI) techniques, such as the two-pool model, are inaccurate and unreliable due to neglecting magnetization transfer (MT) effects and B1 distribution, and the four-pool model is cumbersome with numerous parameters, leading to lengthy analysis times and unclear quantitative values.
Innovation Solution
A medical data processing apparatus and method that utilizes spoiled gradient echo and coherent gradient echo imaging with multiple flip angles to generate low-rank approximate images, incorporating a multi-pool model that includes bound water and free waters with magnetization exchange, using neural networks for high-speed and accurate parameter mapping, including MT effects and B1 distribution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a four-pool model is used to take the MT effect into account, then measurement precision is improved, but device complexity increases due to a large number of parameters
Solution Approach 1:
The patent transforms the complex four-pool model with many parameters into a simplified two-pool model by changing the parameter representation. Instead of directly estimating four separate pools, the invention uses a transformation approach where the four-pool model parameters are converted into two effective pools through mathematical relationships, reducing the number of independent parameters while maintaining measurement accuracy
Solution Approach 2:
The invention extracts and separates the magnetization transfer effect from the standard two-pool model by introducing an additional MT pool. This extraction allows the MT effect to be accounted for independently, enabling accurate measurement of biological tissue structure without requiring the full complexity of a four-pool model
2Measurement precision
If a four-pool model with many parameters is used, then measurement precision is improved, but productivity decreases due to lengthy analysis time
Solution Approach 1:
The patent reduces analysis time by transforming the four-pool model into a two-pool representation, which significantly decreases the computational burden. The parameter transformation allows the system to maintain measurement precision while reducing the number of iterations and calculations required, thereby improving productivity
Solution Approach 2:
The invention performs preliminary transformation of the four-pool model parameters into a two-pool framework before conducting the actual analysis. This preliminary action prepares the data in a simplified format that can be processed more quickly, reducing the overall analysis time while preserving the accuracy benefits of the four-pool model
3Device complexity
If a two-pool model is used, then device complexity is reduced, but measurement precision deteriorates due to neglecting MT effects and B1 distribution
Solution Approach 1:
The patent introduces an intermediary MT pool that mediates between the simple two-pool structure and the complex four-pool model. This intermediary component allows the system to maintain the simplicity of a two-pool model while incorporating the effects of magnetization transfer and B1 distribution through the additional MT pool, thereby improving measurement precision without significantly increasing complexity
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
Enables high-speed and reliable estimation of quantitative parameter maps that accurately reflect biological tissue structure, overcoming the limitations of existing methods by incorporating MT effects and B1 distribution, thereby improving accuracy and reducing analysis time.
Implementation Method 1
there is a problem wherein the two-pool model is inaccurate and unreliable because it does not take into account a magnetization transfer effect (MT effect) or B1 distribution
Implementation Method 2
a two-pool model that takes into account two types of free water
Implementation Method 3
a two-pool model that takes into account two types of free water
Data Source
AI summary
According to one embodiment, a medical data processing apparatus includes processing circuitry. The processing circuitry acquires imaging data acquired by executing spoiled gradient echo imaging and coherent gradient echo imaging by using multiple flip angles. The processing circuitry generates a low-rank approximate image set, which is a set of low-rank approximated images from the imaging data. The processing circuitry reconstructs one or more parameter maps using the above low-rank approximate image set and a related to water exchange in a biological tissue, the multi-pool model including plural free waters and bound water that performs magnetization exchange with those free waters.