Lung MRI Ventilation Imaging via Proton Density Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional lung Fourier decomposition MRI methods provide inaccurate representations of proton density changes associated with ventilation due to contamination from blood volume changes, and lack information on ventilation-dependent lung blood volume, which is crucial for diagnostic pulmonary function assessment.
Innovation Solution
A method that corrects proton density values using geometric information from image registration to separate true ventilation-induced changes from blood volume changes, allowing for the production of accurate ventilation images and ventilation-dependent blood volume images, which are not attainable with conventional methods.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional Fourier decomposition methods are used to analyze lung MRI signals, then the breathing frequency signal can be extracted, but the signal is contaminated by blood volume changes leading to inaccurate ventilation representation
Solution Approach 1:
The patent segments the total proton density change signal into two distinct components: ventilation-induced changes and blood volume changes. By applying image registration to separate geometric transformations from signal intensity changes, the method isolates the true ventilation signal from the confounding blood volume effects, allowing accurate representation of each component independently.
Solution Approach 2:
The patent extracts the blood volume change component from the total signal by using image registration to identify and remove geometric effects. This extraction process isolates the ventilation signal, removing the contaminating blood volume information that would otherwise obscure accurate ventilation measurement.
2Loss of information
If conventional Fourier decomposition is applied to lung MRI, then breathing frequency peaks can be identified, but information about ventilation-dependent blood volume is lost
Solution Approach 1:
The patent segments the signal analysis to separately quantify ventilation and blood volume changes. By dividing the total proton density change into distinct ventilation and blood volume components, the method recovers the previously lost blood volume information while maintaining ventilation measurement accuracy, thereby increasing overall diagnostic information output.
Solution Approach 2:
The patent adds a new dimension to the analysis by separating spatial geometric information from signal intensity information. This dimensional separation allows independent measurement of both ventilation and blood volume changes, transforming a single contaminated measurement into two independent diagnostic parameters.
3Measurement precision
If image registration is used to correct proton density values, then accurate ventilation images can be produced, but the processing complexity increases
Solution Approach 1:
The patent performs image registration as a preliminary step before Fourier analysis. By pre-correcting the images to account for geometric transformations and extracting the ventilation fraction in advance, the method simplifies subsequent signal analysis and reduces the complexity of the main diagnostic workflow.
Solution Approach 2:
The patent introduces image registration as an intermediary process that bridges the raw MRI data and the final Fourier analysis. This intermediary step handles the complex geometric corrections separately, allowing the main signal processing to focus on the simpler task of frequency analysis and ventilation quantification.
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 results in more accurate ventilation images and provides previously unattainable diagnostic information on ventilation-dependent blood volume, enhancing the noninvasive assessment of pulmonary function and potentially aiding in the diagnosis of interstitial lung diseases.
Implementation Method 1
Pulmonary proton MRI has long faced challenges of low signal-to-noise ratio
Implementation Method 2
the time series of each voxel is Fourier transformed. Peaks in the spectrum at the breathing and heart rate frequencies
Data Source
AI summary
A system and method for producing a more accurate ventilation image, as compared to existing lung Fourier decomposition methods, and an image of ventilation dependent blood volume are provided. A time series of images depicting a subject's lungs during free-breathing are acquired and co-registered to a reference image. From the registration process, geometric information indicative proton density changes due to inhalation and exhalation of gas is obtained. This geometric information is used to correct the proton density values in the time series of image frames. These corrected proton density values are Fourier transformed to produce a Fourier spectrum, from which a signal peak occurring at the breathing frequency is extracted and Fourier transformed to produce a more accurate ventilation image. This more accurate ventilation image can be subtracted from a breathing frequency image produced by conventional lung Fourier decomposition methods to produce a ventilation dependent blood volume image.