MRI Field-Map Correction for Metallic Implant Artifacts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current MRI techniques face challenges in generating distortion-free images near metallic implants due to magnetic field inhomogeneity, leading to artifacts such as distortion and blurring, especially in multispectral imaging (MSI) sequences like MAVRIC-SL, where overlapping spectral bins cause shifts and noise sensitivity affects field-map estimation.
Innovation Solution
A novel field-mapping technique using a goodness-of-fit metric and matched-filter to estimate the field-map, combined with an RF-weighted spectral bin combination scheme, to accurately remove bin-specific displacements and improve noise performance, thereby reducing artifacts and achieving a balance between blurring and distortion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional MRI techniques are used to image near metallic implants, then the imaging process is simple and fast, but the images suffer from severe distortion and blurring artifacts due to magnetic field inhomogeneity
Solution Approach 1:
The patent applies preliminary action by performing field-map estimation and correction before final image reconstruction. The method estimates the magnetic field inhomogeneity caused by metallic implants and uses this information to pre-correct the imaging process, thereby reducing distortion and blurring artifacts in the final images
Solution Approach 2:
The patent introduces an intermediary field-map estimation process that acts as a mediator between the raw MRI signals and the final reconstructed images. This field-map serves as intermediate information that characterizes the magnetic field inhomogeneity and enables subsequent correction steps to improve image accuracy
2Reliability
If spectral bin overlap is increased to improve signal coverage, then more complete tissue visualization is achieved, but bin-specific displacements and blurring artifacts increase
Solution Approach 1:
The patent applies feedback by using the estimated field-map to inform and adjust the spectral bin combination process. The field-map estimation provides feedback information about magnetic field inhomogeneity that is used to correct bin-specific displacements and optimize the weighting of different spectral bins, thereby reducing blurring while maintaining signal coverage
Solution Approach 2:
The patent changes parameters by dynamically adjusting the weighting factors applied to different spectral bins based on the estimated field-map. This allows the system to optimize the contribution of each spectral bin to the final image, balancing signal coverage and image sharpness by modifying the combination parameters according to local field inhomogeneity
3Productivity
If center-of-mass field-map estimation is used, then the processing is computationally efficient, but the method is sensitive to noise and produces biased field-map estimates
Solution Approach 1:
The patent introduces an intermediary goodness-of-fit metric that acts as a mediator between the noisy spectral bin signals and the final field-map estimation. This metric provides a more robust basis for estimating field-map values by evaluating how well different field-map hypotheses fit the observed data, thereby reducing noise sensitivity and bias while maintaining computational efficiency
Data Source
AI summary
A method for generating a magnetic resonance image of an object in a magnetic resonance imaging (MRI) system, wherein the object contains at least one metallic implant is provided. The MRI system provides multiple excitations of at least part of the object. The MRI system reads out image signals from the object. The MRI system saves the readout image signals as image data. A field-map is generated from the image data using a goodness-of-fit process which uses a goodness-of-fit metric, matched-filter, and/or similar fitting techniques to fit expected signals from each excitation to the image data.