Quantitative MR Lung Ventilation Mapping via Segmented Registration

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional magnetic resonance tomography (MRT) based lung imaging using Fourier decomposition faces challenges in image registration, signal-to-noise ratio (SNR), and quantification of ventilation maps, particularly in subjects with irregular ventilation, leading to misregistrations and unreliable fractional ventilation calculations.

Innovation Solution

The method involves processing MR lung images by registering them into groups based on ventilation phases, applying a ventilation frequency-based low-pass filter to separate ventilation and perfusion contributions, and using retrospective gating to select regular ventilation images, thereby improving SNR and reducing noise contributions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If direct registration of expiration and inspiration images into medium ventilation position is performed, then quantitative ventilation mapping is achieved, but misregistrations occur due to strong deformation requirements

Engineering Contradiction:
Improveventilation mapping accuracyVSAvoidregistration reliability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent segments the registration process into multiple intermediate steps, introducing virtual ventilation positions between expiration and inspiration. This divides the strong single-step deformation into weaker multi-step transformations, reducing misregistration errors while achieving accurate quantitative ventilation mapping.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If Fourier decomposition is applied to averaged temporal signal intensity, then frequency separation is achieved, but frequency determination becomes unreliable in subjects with irregular ventilation

Engineering Contradiction:
Improvefrequency determination accuracyVSAvoidapplicability to irregular ventilation
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent changes the approach from frequency-domain analysis (Fourier decomposition) to time-domain analysis by detecting ventilation frequency directly from the temporal signal intensity curve. This parameter change enables reliable frequency determination in both regular and irregular ventilation patterns, improving adaptability while maintaining measurement precision.

Inventive Principle:
Principle #35Parameter changes

3Shape

If conventional registration with strong deformation is used, then images are aligned to medium ventilation position, but signal-to-noise ratio decreases due to noise superposition on local lung structures

Engineering Contradiction:
Improveimage alignmentVSAvoidsignal-to-noise ratio
Core Design Contradiction:
ShapeVSObject-affected harmful factors

Solution Approach 1:

By segmenting the registration into multiple steps with intermediate virtual positions, the patent reduces the deformation strength at each step. This prevents noise amplification that occurs with strong single-step deformation, maintaining better signal-to-noise ratio while achieving proper image alignment.

Inventive Principle:
Principle #1Segmentation

4Measurement precision

If extended integration limits are used for frequency separation, then ventilation amplitude calculation is improved, but noise contribution increases

Engineering Contradiction:
Improveventilation amplitude accuracyVSAvoidnoise contribution
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent changes from frequency-domain integration to time-domain frequency detection. This eliminates the need for extended integration limits and their associated noise problems, while still achieving accurate ventilation amplitude calculation through direct temporal signal analysis.

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

This approach enhances the reliability and reproducibility of ventilation maps by providing a direct measure of local ventilation, allowing for more accurate characterization of lung health and diagnosis, while minimizing misregistrations and noise.

Implementation Method 1

magnetic resonance tomography (MRT) based lung imaging using Fourier decomposition

Methodology Applied
Scientific EffectFourier decomposition:

Implementation Method 2

applying a ventilation frequency based low-pass frequency filter on the registered MR lung images

Methodology Applied
Scientific EffectFrequency filtering:

Data Source

PatentEP3107066B1Method of quantitative magnetic resonance lung imaging
Publication Date: 2020.11.11 MEDIZINISCHE HOCHSCHULE HANNOVER
  • EP3107066B1 patent drawingFigure 1
  • EP3107066B1 patent drawingFigure 2A~2C
  • EP3107066B1 patent drawingFigure 3~4

AI summary

A method of processing magnetic resonance (MR) lung images comprises providing a series of the MR lung images collected during multiple ventilation periods, registering the MR lung images for creating registered MR lung images, determining a ventilation frequency from the registered MR lung images, creating a series of ventilation images by applying a ventilation frequency based low-pass frequency filter on the registered MR lung images, and creating a quantitative ventilation map of the lung, which is calculated from the ventilation images, wherein the step of determining the ventilation frequency includes a frequency analysis of a time series of a quantitative lung dimension parameter which is characteristic for the dimension of the lung at the time of collecting the corresponding MR lung image, and the step of creating the quantitative ventilation map includes selecting a first group of ventilation images, which consists of expiration ventilation images collected at regular expiration phases of the multiple ventilation periods, and a second group of ventilation images, which consists of inspiration ventilation images collected at regular inspiration phases of the multiple ventilation periods, and calculating the quantitative ventilation map from the first and second groups of ventilation images, wherein the quantitative ventilation map is adjusted with reference to the tidal volume of the lung.