Combined Correlation Parameter Analysis for MR Relaxometry
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional orthogonal matching pursuit (OMP) techniques for magnetic resonance imaging relaxometry face challenges in accurately quantifying multiple relaxation parameters due to homogeneity and similarity of dictionary entries, leading to inaccurate results and long computation times in solving the inverse problem.
Innovation Solution
A combined correlation approach is employed, where multiple dictionary entries 'vote' to select the most suitable result by averaging correlation values, rather than relying on a single highest correlation, to address the issue of entry homogeneity and similarity, and to improve the accuracy and efficiency of relaxation parameter estimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional OMP techniques use a single highest correlation to select dictionary entries, then the process is simple and fast, but the accuracy of relaxation parameter quantification deteriorates due to homogeneity and similarity of dictionary entries
Solution Approach 1:
The patent combines multiple correlation values from different dictionary entries into a single averaged correlation value. Instead of selecting based on one highest correlation, the method averages correlations across multiple entries that meet a threshold, thereby resolving ambiguities caused by homogeneous dictionary entries and improving quantification accuracy.
Solution Approach 2:
The patent implements a feedback mechanism where the correlation analysis results are used to iteratively refine the selection of dictionary entries. By averaging correlations and using threshold-based filtering, the method feedback-adjusts the selection process to avoid selecting similar entries, thereby improving the distinctiveness and accuracy of parameter quantification.
2Measurement precision
If conventional OMP techniques process multiple relaxation parameters simultaneously, then comprehensive parameter mapping is achieved, but computation time increases due to the large number of dictionary entries
Solution Approach 1:
The patent extracts only the essential correlation information from dictionary entries that meet a threshold criterion, rather than processing all possible combinations of multiple relaxation parameters. By averaging correlations from a selected subset of entries, the method reduces computation time while maintaining comprehensive parameter mapping capability.
Solution Approach 2:
The patent applies partial action by processing only those dictionary entries that exceed a correlation threshold, rather than exhaustively analyzing all entries. This selective approach reduces the computational burden of simultaneous multi-parameter relaxometry while still achieving accurate parameter quantification through the averaged correlation of the selected subset.
3Adaptability or versatility
If dictionary entries are made more similar to cover broader parameter ranges, then adaptability improves, but the ability to distinguish between parameters deteriorates
Solution Approach 1:
The patent changes the parameter used for selection from single highest correlation to averaged correlation across multiple entries. This parameter change allows the system to maintain adaptability across broad parameter ranges while improving parameter distinction capability, because the averaging process highlights entries that are consistently correlated across multiple parameters rather than those that are similar by chance.
Data Source
AI summary
Apparatus, methods, and other embodiments associated with combined correlation parameter estimation are described. One example method includes accessing data associated with a magnetic resonance (MR) signal produced by relaxation of nuclei in an item that has experienced nuclear magnetic resonance (NMR) excitation. The MR signal is a function of two or more NMR parameters. The example method also includes accessing data associated with a set of comparative signal evolutions and computing a value for an NMR parameter based on a combined correlation of the data associated with the MR signal to the data associated with the set of comparative signal evolutions. The combined correlation will depend on at least two correlations between the data associated with the MR signal and two different members of the set of comparative signal evolutions.