MRI Gradient Nonlinearity Correction via Model-Based Reconstruction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current MRI methods suffer from geometric distortion due to gradient nonlinearity, which is not adequately addressed by existing retrospective correction techniques, leading to a tradeoff between geometric accuracy and spatial resolution.

Innovation Solution

A prospective method for correcting gradient nonlinearity during image reconstruction using a model-based estimation, which accounts for errors associated with gradient nonlinearity, thereby reconstructing images while simultaneously correcting for these errors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If retrospective gradient distortion correction (GradWarp) is applied after image reconstruction, then geometric distortion is corrected, but spatial resolution is degraded due to image-domain interpolation

Engineering Contradiction:

Improvegeometric accuracyVSAvoidspatial resolution

Core Design Contradiction:

Manufacturing precisionVSMeasurement precision

Solution Approach 1:

The patent applies preliminary action by correcting gradient nonlinearity effects during the image reconstruction process itself, rather than applying correction after reconstruction. The prospective correction method modifies the reconstruction algorithm to account for gradient deviations from linearity, thereby achieving geometric correction without the need for subsequent image-domain interpolation that degrades spatial resolution

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent replaces the mechanical interpolation-based correction approach (GradWarp) with a model-based estimation approach. Instead of using image-domain interpolation to correct geometric distortion, the method substitutes a prospective correction model that incorporates gradient nonlinearity compensation directly into the reconstruction algorithm, eliminating the need for post-reconstruction processing

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Manufacturing precision

If image-domain interpolation is used for gradient distortion correction, then geometric distortion is corrected, but the correction does not account for finite sampling, undersampling, or noise effects

Engineering Contradiction:

Improvegeometric correctionVSAvoidaccuracy under finite sampling and noise

Core Design Contradiction:

Manufacturing precisionVSReliability

Solution Approach 1:

The patent applies parameter changes by modifying the reconstruction model to include parameters that account for gradient nonlinearity, finite sampling effects, and noise characteristics. The prospective correction method adjusts the reconstruction algorithm's parameters and assumptions to reflect the actual physical constraints of the MRI system, thereby improving the reliability of geometric correction under realistic sampling and noise conditions

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentEP3134747B1Integrated image reconstruction and gradient non-linearity correction for magnetic resonance imaging

Publication Date: 2021.12.01 MAYO FOUNDATION FOR MEDICAL EDUCATION & RESEARCH

Figure 1
Figure 2
Figure 3

AI summary

A system and method for simultaneously reconstructing magnetic resonance images and correcting those imaged for gradient noniinearity effects are provided. As opposed to conventional methods for gradient noniinearity correction where distortion is corrected after image reconstruction is performed, the model-based method described here prospectively accounts for the effects of gradient noniinearity during reconstruction. It is a discovery of the inventors that the method described here can reduce the blurring effect and resolution loss caused by conventional correction algorithms while achieving the same level of geometric correction.