MR Diffusion Parameter Determination via Pre-computed Lookup Tables

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for determining diffusion parameters in MR imaging often converge to local minima rather than global optima, leading to suboptimal results due to the complexity of regression models used.

Innovation Solution

A method that employs an extensive search pattern recognition approach by comparing measured MR signal data with pre-computed or calculated data sets to identify the best matching set, ensuring the determination of globally optimized diffusion parameters.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If regression methods are used to determine diffusion parameters by fitting model functions to measured data, then the calculation process becomes automated and efficient, but the method converges to local minima instead of global optima, leading to suboptimal results

Engineering Contradiction:
Improvecalculation efficiencyVSAvoiddiffusion parameter accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent pre-calculates and stores diffusion parameters for multiple model parameter combinations in a lookup table before actual measurement. This preliminary preparation allows the measurement phase to simply query the pre-computed data, avoiding iterative optimization during actual use and eliminating local minimum problems while maintaining high efficiency.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent creates a comprehensive lookup table that copies pre-calculated diffusion parameters for various model parameter combinations. Instead of performing complex regression analysis on measured data, the system copies the best matching pre-computed parameters from the lookup table based on measured signal characteristics, achieving both speed and accuracy.

Inventive Principle:
Principle #26Copying

2Adaptability or versatility

If complex regression models are used to fit measured MR signals, then the analysis can handle sophisticated diffusion scenarios, but the probability of finding only local minima increases, reducing reliability

Engineering Contradiction:
Improvemodel complexityVSAvoidglobal optimization guarantee
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The patent performs exhaustive model parameter exploration in advance by pre-calculating diffusion parameters for a comprehensive set of model parameter combinations and storing them in a lookup table. This preliminary exhaustive search guarantees finding global optima for all possible scenarios, eliminating the reliability issue of local minima while maintaining adaptability to different diffusion scenarios.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent dynamically adapts to different diffusion scenarios by selecting the best matching pre-computed parameters from the lookup table based on the actual measured signals. The system remains versatile by covering multiple diffusion scenarios in the pre-computed lookup table while ensuring reliable global optimization through the exhaustive nature of the pre-calculation.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP3336572B1Determination of diffusion parameters via a comparison of magnetic resonance measurements with calculated datasets
Publication Date: 2020.09.23 SIEMENS HEALTHCARE GMBH
  • EP3336572B1 patent drawingFigure 1
  • EP3336572B1 patent drawingFigure 2
  • EP3336572B1 patent drawingFigure 3~3A

AI summary

The invention relates to a method for determining at least one diffusion parameter: MR signals are acquired in multiple diffusion measurements, each diffusion measurement being performed with a defined set of MR parameters. The multiple diffusion measurements differ from one another in at least one MR parameter, thereby forming a measured MR signal dataset and the MR signals for the different MR parameters. A plurality of calculated datasets are generated using a model, wherein a defined number of different model parameters are used for all calculated datasets. For each combination of the different model parameters and for different values ​​of the model parameters, an MR signal intensity is calculated for the diffusion measurements.The measured MR signal dataset is compared with the multitude of calculated datasets to identify the calculated dataset that shows the greatest agreement with the measured MR signal dataset based on a quality criterion. The diffusion parameter is derived from the calculated dataset with the greatest agreement.