MRI Motion Correction via Slice-Level State-Space Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional MRI motion correction techniques are limited by their inability to effectively handle inter-slice and intra-slice motion, leading to motion artifacts and requiring sedation or repeated scans, especially in non-cooperative patients like newborns and young children.
Innovation Solution
A slice-level registration-based motion tracking method using a state-space model and outlier-robust Kalman filtering to estimate and correct motion at each slice, allowing for higher temporal resolution and robustness against noise and artifacts, independent of scanner platforms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If volume-level motion correction is used, then the system is simpler to implement, but it cannot effectively handle inter-slice and intra-slice motion with sufficient temporal resolution
Solution Approach 1:
The patent divides the volume-level motion correction problem into slice-level motion correction. Each slice is independently registered and tracked, allowing for more precise correction of inter-slice and intra-slice motion while maintaining manageable system complexity through modular processing of individual slices.
Solution Approach 2:
The patent introduces temporal dimension to the motion correction process by using temporal sequencing of slices and applying Kalman filtering across time. This transforms the spatial-only correction approach into a spatio-temporal approach, improving motion tracking accuracy without proportionally increasing complexity.
2Reliability
If conventional motion correction techniques are used, then the setup is simpler, but they require sedation or repeated scans for non-cooperative patients
Solution Approach 1:
The patent implements feedback through Kalman filtering, which continuously estimates motion parameters from sequential slice data and uses these estimates to correct subsequent slices. This feedback mechanism enables robust motion correction for non-cooperative patients without requiring sedation, as the system adapts to motion patterns in real-time during the scan.
Solution Approach 2:
The motion correction system uses the acquired slice data itself to generate motion correction parameters, eliminating the need for external motion tracking devices or complex setup procedures. The slices self-correct by providing the motion information needed for their own and other slices' correction.
3Speed
If slice-level registration is applied, then temporal resolution for motion tracking is improved, but the processing complexity and computational load increase
Solution Approach 1:
The patent applies preliminary motion estimation using a subset of slices or preliminary registration to establish initial motion parameters before full slice-level registration. This preliminary action reduces the computational burden of subsequent detailed registration while maintaining high temporal resolution for motion tracking.
Solution Approach 2:
The patent uses Kalman filtering to dynamically adjust motion parameters based on temporal patterns in the slice data. By modeling motion as a continuous process with predictable parameter changes over time, the system achieves high temporal resolution without processing every slice at full computational cost.
4Measurement precision
If state-space model with Kalman filtering is used, then robustness to noise and artifacts is improved, but the computational requirements increase
Solution Approach 1:
The patent applies Kalman filtering selectively to estimate motion parameters rather than performing full rigid registration on every slice. This partial action approach provides sufficient motion correction accuracy for most slices while reducing computational energy requirements compared to exhaustive registration methods.
Data Source
AI summary
System and method for processing magnetic resonance imaging (MRI) data of an object to perform motion correction. The method comprises estimating, for each slice of the MRI data, parameters for a state space model using a filter that predicts the location of the object for temporally adjacent slices, wherein the state space model represents motion dynamics of the object throughout acquisition of a three-dimensional volume, registering the slices to a reference image based, at least in part, on the estimated parameters, reconstructing an image based, at least in part, on the registered two-dimensional slices, and outputting the reconstructed image.