PPA Reconstruction Matrix for MRT Calculation Time Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current partial parallel acquisition (PPA) methods in magnetic resonance tomography (MRT) face significant challenges with increased calculation times and hardware requirements due to the quadratic dependency on the number of component coils, leading to unacceptable CPU load and storage demands, especially when using a high number of coils.
Innovation Solution
The method employs an N×M reduction matrix to reduce the number of incomplete data sets, determined through eigenvectors of the covariance matrix or SNR analysis, to accelerate PPA reconstruction, allowing for a subset of incomplete data sets to be reconstructed and transformed into complete image data sets, thereby reducing the number of output channels and calculation time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional PPA reconstruction methods are used with a high number of component coils, then complete spatial coding and image reconstruction are achieved, but calculation time and CPU load increase quadratically
Solution Approach 1:
The patent extracts and processes only a subset of incomplete data sets (M out of N) rather than all data sets. By selecting a representative subset for reconstruction, the calculation time is reduced while maintaining adequate image quality, thus resolving the contradiction between complete reconstruction and calculation time.
Solution Approach 2:
Instead of performing complete reconstruction on all N data sets, the patent applies partial action by reconstructing only M data sets where M < N. This partial reconstruction approach reduces computational load quadratically while still providing sufficient spatial coding information for acceptable image quality.
2Measurement precision
If a high number of component coils are used in PPA acquisition, then spatial coding capability and SNR are improved, but hardware requirements and storage demands increase
Solution Approach 1:
The patent extracts only the necessary subset of data sets (M out of N) for reconstruction purposes. This reduces the storage requirements and hardware processing demands while maintaining the SNR benefits of having multiple coils, as the subset still provides sufficient spatial information.
Solution Approach 2:
By using partial action (processing only M data sets instead of all N), the patent reduces hardware resource requirements and storage demands while retaining adequate SNR performance from the coil array configuration.
3Reliability
If all incomplete data sets are reconstructed using conventional PPA methods, then complete image data sets are obtained, but the reconstruction process becomes computationally prohibitive
Solution Approach 1:
The patent extracts a subset of M incomplete data sets from the total N data sets for reconstruction. This extraction approach maintains adequate image completeness by selecting representative data sets while dramatically improving reconstruction efficiency by avoiding processing of all N data sets.
Solution Approach 2:
The patent applies partial action by reconstructing only M data sets instead of all N data sets. This partial reconstruction maintains sufficient image completeness for diagnostic purposes while significantly improving reconstruction efficiency and reducing computational burden.
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 significantly reduces the calculation time for GRAPPA reconstruction while maintaining a slight increase in SNR and minimal under-sampling artifacts, even with a high number of component coils, by forming a subset of incomplete data sets and reconstructing only those, rather than completing all data sets, thus optimizing image reconstruction efficiency.
Implementation Method 1
MRT is based on the physical phenomenon of magnetic resonance and has been successfully used as an imaging method for over 15 years in medicine and biophysics. In this examination modality, the subject is exposed to a strong, constant magnetic field. The nuclear spins of the atoms in the subject, which were previously randomly oriented, thereby align.
Implementation Method 2
By the use of inhomogeneous magnetic fields generated by gradient coils, the measurement subject can be spatially coded in all three spatial directions.
Data Source
AI summary
In a method and apparatus for generating a magnetic resonance image of a contiguous region of a human body on the basis of partial parallel acquisition (PPA) by excitation of nuclear spins and measurement of radio-frequency signals indicating the excited spins, the spin excitation is implemented in steps with measurement of an RF response signal simultaneously in each of a number of N component coils. A number of response signals thus are acquired that, for each component coil, form an incomplete data set (40) of acquired RF signals. Additional acquired calibration data points exist for each incomplete data set. The N incomplete data sets are acquired to a subset of M reduced, incomplete data sets on the basis of an N×M reduction matrix, so that M reduced, incomplete data sets are obtained, M complete data sets are formed on the basis of an N×M reconstruction matrix with the non-measured lines of the M reduced, incomplete data sets being reconstructed from all N incomplete data sets. A spatial transformation of the completed reduced data sets is then implemented in order to form a complete image data set from each completed, reduced data set.