fMRI Reconstruction via Compressed Sensing and Radial Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current fMRI technologies face challenges in achieving high image resolution and frame rate due to the need for a large number of k-space views, which increases scan time, and often result in streak artifacts from insufficient sampling, especially in three-dimensional imaging.
Innovation Solution
The method involves acquiring highly undersampled image frames with interleaved views, using a priori knowledge of the NMR signal contour to weight signal distribution in the backprojection process, and reconstructing images using a composite image to enhance signal-to-noise ratio (SNR) and reduce artifacts, employing a 3D hybrid projection reconstruction pulse sequence with radial trajectories and phase encoding for multiple slices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a large number of k-space views are acquired, then image quality is improved, but scan time increases
Solution Approach 1:
The patent applies partial action by acquiring only a subset of k-space views (e.g., 10-20 radial views instead of full Fourier coverage) and using compressed sensing algorithms to reconstruct images from this incomplete data. This reduces scan time while maintaining acceptable image quality through intelligent reconstruction rather than complete data collection.
Solution Approach 2:
The patent replaces the traditional mechanical approach of systematically scanning through all k-space views with a computational approach using compressed sensing and iterative reconstruction algorithms. This substitution allows image reconstruction from highly undersampled data, reducing the physical scanning time while maintaining image quality through mathematical optimization.
2Loss of time
If the number of acquired views is reduced, then scan time is reduced, but streak artifacts are produced
Solution Approach 1:
The patent replaces traditional Fourier transform-based reconstruction with compressed sensing algorithms that can handle incomplete and irregularly sampled data. This computational approach eliminates streak artifacts caused by insufficient sampling by using sparsity constraints and iterative optimization to reconstruct images from highly undersampled k-space data.
Solution Approach 2:
The patent changes the reconstruction parameters and algorithms to accommodate highly undersampled data. By using compressed sensing with appropriate regularization parameters and iterative reconstruction methods, the system can produce artifact-free images from data that would traditionally produce severe streak artifacts.
3Speed
If radial projection reconstruction is used, then frame rate is improved, but insufficient sampling causes artifacts
Solution Approach 1:
The patent performs preliminary actions by acquiring a small number of radial k-space views at high frame rates and then using compressed sensing algorithms to reconstruct the images. The preliminary radial sampling captures the essential signal information, and the computational reconstruction fills in the missing data without artifacts, enabling high frame rate imaging with adequate sampling.
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 allows for the production of high-quality fMRI images with significantly fewer acquired views, reducing artifacts and increasing SNR, while maintaining or improving image resolution and frame rate, thus optimizing fMRI data acquisition and reconstruction.
Implementation Method 1
magnetic field gradients (Gx, Gy and Gz) are employed. Typically, the region to be imaged is scanned by a sequence of measurement cycles in which these gradients vary according to the particular localization method being used
Implementation Method 2
the individual magnetic moments of the spins in the tissue attempt to align with this polarizing field, but precess about it in random order at their characteristic Larmor frequency
Implementation Method 3
If the substance, or tissue, is subjected to a magnetic field (excitation field B1) which is in the x-y plane and which is near the Larmor frequency, the net aligned moment, Mz, may be rotated, or 'tipped', into the x-y plane to produce a net transverse magnetic moment Mt
Implementation Method 4
A signal is emitted by the excited spins after the excitation signal B1 is terminated, this signal may be received and processed to form an image
Data Source
AI summary
Acquisition of MR data during a fMRI study employs a hybrid PR pulse sequence to acquire projection views from which multi-slice image frames may be reconstructed that depict the BOLD response to an applied stimulus or performed task. Composite images are reconstructed at each slice using the combined interleaved projection views from all the acquired image frames. The composite images are used to reconstruct the highly undersampled image frames.