Synthetic MRI Image Generation with Switchable Contrast Components

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current quantitative Magnetic Resonance Imaging (MRI) techniques struggle to generate a wide range of image contrasts, particularly those involving additional weighting mechanisms like Magnetization Transfer (MT) and fat attenuation, due to their impact on entire images and signal-to-noise ratio, limiting the clinical adoption of synthetic image generation.

Innovation Solution

A method and system for generating synthetic images with switchable contrast components, using multiple quantitative MRI acquisitions to create proton-density images with different weightings, allowing users to interactively turn contrast components on or off through a user interface, thereby enabling a variety of contrasts based on short acquisition times.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If preparation modules (e.g., fat saturation, magnetization transfer) are used during qMRI acquisition to generate additional contrast weighting, then the range of image contrasts is extended, but the signal-to-noise ratio deteriorates and acquisition time increases

Engineering Contradiction:
Improverange of image contrastsVSAvoidsignal-to-noise ratio
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The patent applies preliminary action by acquiring multiple quantitative parameter maps (T1, T2, T2*) during the initial qMRI acquisition without applying preparation modules. These quantitative maps are then used in a signal model to synthetically generate various contrast-weighted images (including fat-saturated and MT-weighted contrasts) post-acquisition, thereby avoiding the need to apply preparation modules during the actual contrast-generating acquisitions.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent uses copying by creating synthetic replicas of contrast-weighted images through signal modeling. Instead of directly acquiring images with different preparation modules, the system copies the underlying tissue properties from the quantitative maps and reconstructs multiple contrast variants computationally, preserving signal-to-noise ratio while providing diverse contrast options.

Inventive Principle:
Principle #26Copying

2Adaptability or versatility

If multiple conventional acquisitions are performed to obtain different contrast-weighted images, then the range of image contrasts is improved, but the acquisition time increases

Engineering Contradiction:
Improverange of image contrastsVSAvoidacquisition time
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent implements multi-functionality by designing a single qMRI acquisition sequence that simultaneously collects data for multiple quantitative parameters (T1, T2, T2* maps). This unified acquisition serves multiple purposes: it provides the foundational quantitative information needed to generate any desired contrast-weighted image through signal modeling, eliminating the need for separate dedicated acquisitions for each contrast type.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent applies parameter changes by varying the input parameters to the signal model (different sequence parameters TE, TR, TI combined with the measured quantitative tissue parameters) to generate different contrast-weighted images from the same acquired data, allowing multiple contrasts to be derived from a single acquisition set.

Inventive Principle:
Principle #35Parameter changes

3Adaptability or versatility

If TI is adjusted to suppress fat signal in synthetic images, then fat suppression is achieved, but the entire image contrast is altered and signal-to-noise ratio decreases

Engineering Contradiction:
Improvefat suppression capabilityVSAvoidimage contrast control
Core Design Contradiction:
Adaptability or versatilityVSEase of operation

Solution Approach 1:

The patent applies local quality by enabling independent control of fat suppression for specific tissue types while preserving the original image contrast characteristics for other tissues. The system selectively applies fat saturation modeling only to fat tissue based on its identified location and properties from the quantitative maps, leaving the contrast weighting for other tissues unchanged.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent extracts the fat signal component from the overall image using the quantitative fat tissue identification from T1 and T2 maps. By separating the fat contribution, the system can selectively suppress or modify only the fat signal while maintaining the original contrast characteristics for non-fat tissues, rather than altering the entire image contrast.

Inventive Principle:
Principle #2Taking out (Extraction)

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

This approach provides a large variety of contrasts with improved diagnostic capabilities by allowing radiologists to change contrasts post-acquisition and switch between different effects, such as fat suppression and MT-weighting, enhancing the diagnostic process.

Implementation Method 1

Quantitative MRI (qMRI) is becoming a well-established technique especially in musculoskeletal (MSK) research

Methodology Applied
Scientific EffectMagnetic Resonance: Magnetic Field

Implementation Method 2

The image contrast of the synthetic image S, usually called 'synthetic contrast', depends on sequence parameters (inversion time TI, repetition time TR, echo time TE) and tissue properties (longitudinal relaxation T1, transverse relaxation T2, and initial magnetization M0)

Methodology Applied
Scientific EffectSignal relaxation:

Data Source

PatentEP3726461B1Method and system for generating synthethic images with switchable image contrasts
Publication Date: 2021.10.27 SIEMENS HEALTHCARE GMBH
  • EP3726461B1 patent drawingFigure 1
  • EP3726461B1 patent drawingFigure 2
  • EP3726461B1 patent drawing

AI summary

The present invention concerns a system and a method for generating a synthetic image with switchable image contrast components for a biological object, the method comprising: - using a first quantitative MRI acquisition technique (101) for measuring a value of a first quantitative parameter Q1 for the biological object and generating a first quantitative map of the biological object, wherein said first quantitative MRI acquisition technique is configured for generating at least a first contrast-weighted image for said biological object; - using a second quantitative MRI acquisition technique (102) for measuring a value of a second quantitative parameter Q2 for the biological object and generating a second quantitative map of the biological object comprising the values of Q2 for said biological object, wherein said second quantitative MRI acquisition technique is configured for generating at least a second contrast-weighted image for said biological object; - using (103) the obtained first and second quantitative maps, and the first and second contrast weighted images as inputs in a physical signal model configured for generating a synthetic image M with arbitrary sequence parameters P1, P2, P3, according to: M=CifQ1Q2P1P2P3 wherein Ci with i=1,2, is a contrast component for the generation of the synthetic image M coming from respectively the first contrast-weighted image (i=1) and the second contrast-weighted image (i=2); f is a function of Q1, Q2, P1, P2 and P3; - displaying the synthetic image M generated by using one of the contrast components Ci, wherein an activation of a contrast switch is configured for automatically displaying the synthetic image M generated by using another one of the contrast components Ci.