Parallel Imaging Parameter Optimization in MRI
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In 3D magnetic resonance imaging with parallel imaging, selecting appropriate acceleration factors and encoding directions is complex, often resulting in poor image quality due to incorrect choices, and involves a cumbersome setup prone to operator mistakes, with optimal selection requiring prior scans for noise information.
Innovation Solution
An automatic optimization system for parallel acceleration parameters that simplifies the selection of encoding directions and reduction factors based on quantifiable parameters, including minimum signal-to-noise ratio (SNR), using coil sensitivity information and patient loading data from a conventional coil survey scan to optimize encoding directions and reduction factors, reducing the number of parameters to be selected and streamlining the user interface.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual selection of acceleration factors and encoding directions is used, then operator control over scan parameters is maintained, but the system becomes complex and prone to operator mistakes
Solution Approach 1:
The system automatically determines optimal acceleration factors and encoding directions by utilizing coil sensitivity information and noise covariance data from the scan itself, eliminating the need for manual parameter selection and operator intervention in the complex setup process
Solution Approach 2:
The system performs a preliminary survey scan to acquire coil sensitivity information and noise covariance data before the actual diagnostic scan, using this pre-acquired information to automatically optimize the acceleration parameters for the main scan
2Productivity
If multiple individual acceleration factors are selected for different encoding directions, then optimized parallel imaging performance can be achieved, but the setup becomes complicated and time-consuming
Solution Approach 1:
The system automatically determines the optimal distribution of acceleration factors across different encoding directions by analyzing coil sensitivity information and noise covariance, eliminating the need for manual configuration of multiple individual acceleration factors and significantly reducing setup time
Solution Approach 2:
The system performs a preliminary survey scan to acquire the necessary coil sensitivity and noise information, which is then used to automatically optimize the acceleration factor distribution for the main diagnostic scan, reducing configuration time for the actual imaging
3Loss of time
If incorrect acceleration factors are selected, then scan time is reduced, but image quality deteriorates due to low signal-to-noise ratio
Solution Approach 1:
The system uses noise covariance data acquired during the survey scan to predict the signal-to-noise ratio for different acceleration factor combinations, providing feedback to automatically select the optimal acceleration factors that maintain image quality while reducing scan time
Solution Approach 2:
The system changes the acceleration factors based on the specific coil configuration and noise characteristics detected during the survey scan, automatically optimizing the balance between scan time reduction and image quality for each specific imaging scenario
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A parallel magnetic resonance imaging system (1) includes at least one radio frequency (RF) coil (10, 12) with a plurality of coil elements, a smart select unit (24), a parallel imaging parameter unit (28), and a sequence control (16). The smart select unit (24), from a pre-scan or prior scan of a subject with the at least one RF coil, constructs (60) a signal map and a plurality of noise maps based on different sets of reduction factors. The parallel imaging parameter unit (28) selects a set of reduction factors corresponding to a noise map which includes a highest signal-to-noise ratio (SNR). The sequence control (16) performs a magnetic resonance imaging scan of the subject based on the selected reduction factors.