MRI Reconstruction Using Phase-Corrected K-Space Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current MRI methods face challenges in reconstructing magnetic resonance images from incomplete data records without artifacts, particularly due to phase differences between segments, which often require additional navigators or lose phase and coil profile information during reconstruction.
Innovation Solution
A method involving the calculation of a first reconstruction kernel from a reference spectrum, preliminary reconstruction of MR segment data records, phase image calculation, modified reference spectra creation, and subsequent second reconstruction kernels to reconstruct missing data without phase artifacts, allowing for artifact-free image reconstruction without additional navigator measurements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If parallel imaging with incomplete k-space sampling is used to reduce acquisition time, then productivity is improved, but measurement precision deteriorates due to missing data and reconstruction artifacts
Solution Approach 1:
The method performs preliminary reconstruction of each MR segment data record to obtain preliminary phase information before final image reconstruction. This preliminary action enables the calculation of phase correction factors that are then applied during the final reconstruction, thereby resolving phase artifacts while maintaining the benefits of accelerated parallel imaging
Solution Approach 2:
The method uses the preliminary reconstructed phase information as feedback to correct phase errors in the final reconstruction process. By iteratively using the preliminary results to inform the final reconstruction, the method eliminates phase artifacts and improves measurement precision without sacrificing the accelerated acquisition time
2Measurement precision
If traditional reconstruction methods are used to handle phase differences between segments, then measurement precision is improved, but device complexity increases due to additional navigator measurements
Solution Approach 1:
The method enables the MR segment data records to self-correct their phase errors through the preliminary reconstruction process. Each segment's preliminary phase information is used to generate its own correction factors, eliminating the need for external navigator measurements or complex inter-segment coordination, thereby reducing device complexity while maintaining measurement precision
3Productivity
If conventional parallel imaging reconstruction is applied, then productivity is improved, but loss of information occurs due to phase and coil profile information loss
Solution Approach 1:
The method performs preliminary reconstruction to extract and preserve phase information before the final reconstruction step. This preliminary action ensures that phase and coil profile information is captured and utilized in the final reconstruction, preventing information loss while maintaining fast reconstruction speeds through efficient parallel processing
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 enables the reconstruction of magnetic resonance images from undersampled data records with reduced phase artifacts, improving image quality by incorporating continuous phase information and maintaining coil profile details, thus enhancing the accuracy and reliability of MRI reconstructions.
Implementation Method 1
nuclear spins contained in a measurement volume of the examination object are oriented with respect to the direction of the main magnetic field... nuclear spins... carry out a precession movement as a result of being irradiated by electromagnetic RF pulses... MR signals are recorded
Implementation Method 2
By virtue of the object region being exposed to a substantially static and homogeneous main magnetic field in a magnetic resonance measuring apparatus, nuclear spins contained in a measurement volume of the examination object are oriented with respect to the direction of the main magnetic field
Implementation Method 3
this magnetization within the measurement volume is excited to carry out a precession movement as a result of being irradiated by electromagnetic RF pulses (RF: radiofrequency), the frequency of said precession movement being proportional to the local magnetic field strength
Data Source
AI summary
A method for generating a magnetic resonance image includes providing MR segment data records, wherein each MR segment data record has N×M frequency voxels in k-space. To reduce the acquisition time during MR segment recordings, the amount of MR data is reduced by incompletely sampling the k-space during a recording. The missing data of the MR segment data records are reconstructed twice: Preliminarily reconstructed MR segment data records are calculated first, with a reconstruction kernel obtained from reference data. Modified reference images containing phase information are obtained by creating phase images from the preliminarily reconstructed MR segment data records and combining these phase images with the absolute value of the reference image generated from the reference data. The second reconstruction kernels are ascertained therefrom in turn. In contrast to the first reconstruction kernel, these contain phase information, such that the missing data can be reconstructed without phase artifacts.