MRI Shimming Using Machine Learning to Reduce Scan Time
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Object-specific shimming in MRI systems is a time-consuming process due to the need for separate shimming scans to correct magnetic field inhomogeneities, which affects image quality and signal-to-noise ratio, especially in applications sensitive to small chemical shifts like spectral fat saturation and water excitation.
Innovation Solution
A method utilizing a trained machine learning module within the MRI system to generate shimming information from pre-scan data, excluding B1 maps, which directly determines shim currents or B0 inhomogeneity maps to correct the magnetic field, thereby eliminating the need for a separate shimming scan.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a separate shimming scan is performed to correct object-specific magnetic field inhomogeneities, then the magnetic field homogeneity is improved, but the total scanning time increases
Solution Approach 1:
The patent combines the shimming scan with the main imaging scan by using the same acquired data for both purposes. The pre-scan data obtained for imaging purposes is reused to determine B0 inhomogeneities and calculate shim parameters, eliminating the need for a separate dedicated shimming scan while maintaining field homogeneity correction.
Solution Approach 2:
The patent makes the pre-scan data serve multiple functions: it is used both for generating the main MR image and for determining the B0 inhomogeneities required for shimming. This multi-functional use of the same data acquisition removes the time penalty associated with separate shimming scans.
2Measurement precision
If traditional separate shimming scans are used to estimate B0 inhomogeneities, then accurate shimming parameters are obtained, but the process becomes time-consuming
Solution Approach 1:
The patent performs the shimming-related measurements as a preliminary action during the main scan acquisition. By extracting B0 inhomogeneity information from the pre-scan data that is already being acquired for imaging purposes, the system obtains accurate field maps without requiring additional time for separate measurements.
3Manufacturing precision
If object-specific shimming is performed with separate scans, then image quality is improved, but the workflow complexity increases
Solution Approach 1:
The patent merges the shimming workflow into the main imaging workflow by using the same pre-scan data for both image generation and shimming parameter calculation. This integration simplifies the overall process by eliminating separate scanning steps and reducing the number of operations the operator must manage.
Data Source
AI summary
Object specific in-homogeneities in an MRI system are corrected. Prescan information available at the MR imaging system is determined. The prescan information includes at least object specific information of an object located in the MR imaging system from which an MR image is to be generated. The prescan information does not include a B1 map of the MRI system with the object being present in the MR imaging system. The prescan information is applied to a trained machine learning module provided at the MRI system. The trained machine learning module determines and generates shimming information as output. The shimming information is applied to a shimming module of the MR imaging system, wherein the shimming module uses the shimming information to generate a corrected magnetic field B0.