MRI Signal Reconstruction via Coupled System Inversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current MRI image reconstruction methods struggle with correcting artifacts caused by patient movement during acquisition, requiring prior knowledge of movement parameters and being limited to affine transformations, which restricts their applicability to complex movements and increases the need for calibration steps.
Innovation Solution
A method that uses a coupled system inversion (GRICS) to reconstruct MRI images by modeling both the acquisition chain and motion disturbances directly from corrupted data, eliminating the need for initial calibration and allowing for the correction of complex movements through a parametric model of elastic physiological movements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If prior art methods use external sensors to provide movement information for correction, then movement correction capability is improved, but device complexity and calibration requirements increase
Solution Approach 1:
The system uses the corrupted MRI data itself to estimate the movement parameters through an optimization process that minimizes a cost function. The method is self-contained and does not require external sensors or separate calibration procedures, as the movement information is extracted directly from the data being reconstructed.
Solution Approach 2:
The patent introduces a cost function as an intermediary that bridges the corrupted data and the movement parameters. This cost function quantifies the inconsistency caused by movement and guides the optimization process to estimate parameters without requiring external measurement devices or calibration data.
2Ease of manufacture
If prior art methods assume affine transformations for movement, then mathematical tractability is improved, but adaptability to complex movements deteriorates
Solution Approach 1:
The method transitions from static affine transformation models to dynamic, data-driven parameter estimation. The optimization process adapts to the actual movement patterns present in the data, allowing the system to handle complex non-affine movements while maintaining mathematical tractability through iterative refinement of transformation parameters.
Solution Approach 2:
The patent changes the parameters being estimated from fixed affine transformation parameters to time-varying parameters that can capture complex movement patterns. The optimization process adjusts these parameters iteratively to minimize the cost function, enabling adaptation to diverse and complex movement types without requiring a predetermined transformation model.
3Object-generated harmful factors
If prior art methods delete corrupted data and use parallel imaging, then artifact reduction is improved, but information loss increases
Solution Approach 1:
Instead of discarding corrupted data, the method converts the harmful effect of movement into useful information by using the corrupted data itself to estimate movement parameters. The inconsistencies caused by movement are transformed into constraints that guide the reconstruction process, allowing full utilization of the acquired data while correcting for motion effects.
Solution Approach 2:
The optimization process implements a feedback mechanism where the reconstructed image quality and data consistency are continuously evaluated against the cost function. This feedback guides iterative refinement of the movement parameter estimates and image reconstruction, allowing the system to recover information that would be lost in simple deletion approaches.
4Measurement precision
If prior art methods repeat acquisitions to ensure data quality, then measurement precision is improved, but acquisition time increases
Solution Approach 1:
The method maintains continuous acquisition without interruption for quality checks or re-acquisitions. The corrupted data from physiological movements is continuously processed through the optimization framework, which extracts useful information throughout the entire acquisition period, eliminating the need for repeated measurements while maintaining data quality.
Solution Approach 2:
The patent changes the approach from repeating acquisitions with fixed parameters to a single continuous acquisition with dynamically optimized parameters. The optimization process adjusts movement parameter estimates in real-time during reconstruction, allowing the system to achieve high measurement precision without extending the acquisition time through multiple passes.
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 effectively reduces the impact of motion disturbances on image reconstruction, enabling the correction of complex movements without prior knowledge of movement parameters, thus improving image quality and reducing artifacts.
Implementation Method 1
Magnetic resonance imaging (MRI) is a non-invasive radiological technique based on the physical phenomenon of nuclear magnetic resonance (NMR)
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a method for acquiring (1) experimental measures with interferences of a physical phenomenon, and for reconstructing (2) a point-by-point signal (3) representative of said phenomenon according to at least one dimension that can vary during the experimental measure acquisition, using at least one simulation model (4) of at least one acquisition chain of said experimental measures including at least one interference, and at least one model (8) of each interference in each acquisition chain, each interference model (8) being determined at least from the measures themselves, characterised in that the simulation and interference models include adjustable parameters (6, 10) depending on experimental conditions, wherein at least one adjustable parameter of one of said models is coupled to at least one adjustable parameter of the other model, and in that the adjustable parameters (6, 10) are optimised (2) in a coupled manner. The invention also relates to a device for MRI imaging, NMR, or medical imaging using such a method.