Magnetic Resonance Image Reconstruction Using Calibration Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Magnetic resonance imaging in medicine faces challenges with the duration required for acquiring data, particularly in parallel imaging techniques like SMS imaging, where slice crosstalk artifacts can occur due to insufficient slice separation, leading to reduced image quality and noise.
Innovation Solution
Incorporating calibration data from a reference scan into the final magnetic resonance image data set, using similar recording parameters for both, to enhance image quality and reduce artifacts by increasing the signal-to-noise ratio and eliminating slice acceleration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If slice acceleration is applied in SMS imaging to reduce data acquisition time, then productivity is improved, but slice crosstalk artifacts increase and image quality deteriorates
Solution Approach 1:
The calibration data set is acquired in advance during a reference scan before the actual imaging process. This preliminary acquisition of calibration data with high signal-to-noise ratio allows the separation algorithm to properly distinguish between simultaneously acquired slices, preventing slice crosstalk artifacts while maintaining the acceleration benefits of SMS imaging
Solution Approach 2:
The calibration data set acts as an intermediary that mediates between the accelerated multi-slice data and the final reconstructed images. By using this intermediate calibration data in the separation algorithm, the system can resolve the conflict between fast acquisition and high image quality, as the calibration data provides the necessary reference information for accurate slice separation without requiring additional acceleration-related compromises
2Loss of time
If multiple slices are recorded simultaneously in SMS imaging, then data acquisition time is reduced, but slice separation becomes insufficient leading to crosstalk artifacts
Solution Approach 1:
The system performs preliminary acquisition of calibration data in a reference scan before the actual accelerated imaging. This advance preparation ensures that high-quality reference information is available to support the separation algorithm when multiple slices are simultaneously acquired, maintaining reliable slice separation despite the time constraints of accelerated imaging
Solution Approach 2:
The calibration data set serves as a copy or reference model that captures the spatial and signal characteristics of the imaging scenario. This copied reference information is then used by the separation algorithm to accurately disentangle the simultaneously acquired slices, ensuring reliable slice separation without requiring the actual imaging process to be slower
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 improves image quality by increasing the signal-to-noise ratio and significantly reduces slice crosstalk artifacts, while potentially shortening data acquisition time by omitting unnecessary recording passes.
Implementation Method 1
The invention concerns a method for generating a magnetic resonance image data set of a target region using a magnetic resonance scanner
Data Source
AI summary
In a method and apparatus for generating a magnetic resonance (MR) image data set of a target region, MR data for a first number of slices are recorded and the recording of MR data for a second number, which is smaller than or equal to the first number, of different slices takes place simultaneously. A separation algorithm of the parallel imaging is used to determine MR data that are assigned to individual slices from the multi-slice data set produced during the simultaneous recording of the multiple slices. This separation algorithm uses input parameters determined from a calibration data set of the target region, the calibration data set being recorded in a reference scan, after which the MR image data set is reconstructed from the MR data assigned to individual slices, wherein at least part of the calibration data set is also used for reconstructing the MR image data set.