Quasi-diffusion MRI using constrained CTRW model

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current diffusion imaging techniques, such as Diffusional Kurtosis Imaging (DKI) and diffusion tensor imaging (DTI), face challenges in accurately representing physical diffusion processes, leading to unstable parameterization, inefficient computation, and suboptimal utilization of modern MRI systems.

Innovation Solution

A computer-implemented method using a constrained continuous-time random walk (CTRW) model, where the time and space parameters are related by a correlation function, reducing the degrees of freedom and enabling faster data acquisition with lower noise and higher accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If DKI uses a simple parameterisation of the diffusion signal, then the technique can be implemented, but it does not accurately represent actual physical diffusion processes and cannot utilise diffusion sensitisations for which b>3000 s mm -2

Engineering Contradiction:
Improveaccuracy of diffusion parameter representationVSAvoidutilisation of high diffusion sensitisation capabilities
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent changes the parameterisation approach from simple DKI to a continuous-time random walk (CTRW) model with fractional diffusion, allowing accurate representation of non-Gaussian diffusion while enabling utilisation of high b-value data (b>3000 s mm -2) that DKI cannot handle

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If geometrically constrained diffusion models are used, then potentially useful clinical information can be provided, but the models are challenging to fit and exhibit acquisition and parameter based bias that limit their application

Engineering Contradiction:
Improveclinical information accuracyVSAvoidmodel fitting complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces the mechanical/geometric constraint-based models (which require complex fitting and have acquisition biases) with a stochastic CTRW model that naturally accounts for diffusion heterogeneity without requiring complex geometric assumptions, thereby reducing fitting complexity and bias

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Measurement precision

If standard diffusion imaging techniques are used, then diffusion parameters can be computed, but the computed images are very susceptible to noise, leading to the requirement for long acquisition times

Engineering Contradiction:
Improvediffusion parameter accuracyVSAvoidacquisition time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent changes the diffusion signal parameterisation to use CTRW model parameters (diffusion coefficient D and fractional order α) that are more robust to noise, enabling accurate diffusion parameter computation with shorter acquisition times while reducing susceptibility to noise

Inventive Principle:
Principle #35Parameter changes

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

The method allows for rapid image acquisition with high signal-to-noise and contrast-to-noise ratios, overcoming limitations of existing techniques and providing accurate diffusion parameter maps that can be used for clinical diagnosis and research.

Implementation Method 1

Nuclear magnetic resonance (NMR) is a well-known and powerful tool for analysing the properties and structure of materials

Methodology Applied
Scientific EffectNuclear magnetic resonance: Magnetic Field

Data Source

PatentEP3997473B1A method and apparatus for quasi-diffusion magnetic resonance imaging
Publication Date: 2025.06.04 CITY
  • EP3997473B1 patent drawingFigure 1(a)~2
  • EP3997473B1 patent drawingFigure 3(i)~3(ii)
  • EP3997473B1 patent drawingFigure 4

AI summary

A computer-implemented method of analysing nuclear magnetic resonance, NMR, data of a target object is provided. The method comprises receiving NMR data of the target object, and analysing the received NMR data using a model of the diffusive behaviour of 5 particles within the target object. The model includes a time parameter and a space parameter, the time parameter describing temporal characteristics of the diffusive behaviour of particles in the model and the space parameter describing spatial characteristics of the diffusive behaviour of particles in the model. The model is constrained such that the value of the time parameter and the value of the space parameter 10 are related according to a correlation function. An apparatus for analysing nuclear magnetic resonance, NMR, data of a target object is also provided.