Multi-Shot MRI Reconstruction With Joint Phase Error Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for reconstructing multi-shot magnetic resonance imaging (MRI) data assume smooth relative phase differences between shots, which are not always satisfied, leading to image artifacts and inefficiencies in reconstruction.
Innovation Solution
A method for reconstructing MRI images using a computer system that incorporates shot-dependent phase data and iteratively updates the phase data based on self-consistency, combining phase-aware image reconstruction with data consistency-based phase estimation, and optionally using machine learning priors to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If low-rank representation with smooth phase assumption is used for multi-shot reconstruction, then reconstruction speed is improved, but image accuracy deteriorates due to artifacts when the assumption is not satisfied
Solution Approach 1:
The patent transitions from static phase assumptions to dynamic phase estimation by iteratively updating shot-dependent phase values during reconstruction. The phase parameters are no longer fixed but adaptively refined through multiple iterations, allowing the system to handle non-smooth phase variations while maintaining computational efficiency.
Solution Approach 2:
The reconstruction algorithm performs self-calibration by automatically estimating and correcting shot-dependent phase errors without requiring external calibration data or manual intervention. The method uses the acquired multi-shot data itself to derive and refine phase corrections iteratively, making the system self-sufficient.
2Device complexity
If traditional parallel imaging reconstruction is used without phase correction, then device complexity is reduced, but image quality deteriorates due to phase-related artifacts
Solution Approach 1:
The patent merges parallel imaging reconstruction with phase estimation and correction into a unified iterative framework. Instead of separate processing steps, the method jointly optimizes image reconstruction and phase correction, allowing both functions to inform and enhance each other through shared computational resources and iterative refinement.
Solution Approach 2:
The method introduces shot-dependent phase parameters that are iteratively updated during reconstruction. These phase parameters are dynamically adjusted based on data consistency checks, allowing the system to adapt to actual phase variations in the multi-shot data without requiring complex pre-calibration procedures.
3Manufacturing precision
If iterative phase update based on self-consistency is implemented, then image accuracy is improved, but computational time increases
Solution Approach 1:
The patent implements a feedback mechanism where the reconstructed images are used to update phase estimates, which in turn improve subsequent reconstructions. This closed-loop approach continuously refines both image quality and phase accuracy, with each iteration building upon previous results to converge toward an optimal solution.
Solution Approach 2:
The method performs preliminary phase estimation using initial reconstructed images before refining the reconstruction with corrected phase values. This two-stage approach allows the system to quickly establish a baseline and then progressively improve accuracy, balancing computational effort with image quality gains.
Data Source
AI summary
Images are reconstructed from k-space acquired with a magnetic resonance imaging (“MRI”) system using a multi-shot pulse sequence. In each iteration, a phase-aware image reconstruction, a data-consistency update across all shots or subsets of data, and a relative phase estimation across the reconstructed images for each shot are performed. In this way, the reconstruction framework recasts the problem as an iterative relative phase estimation problem, which allows for the use relative phase estimation techniques. Through an iterative search, an artifact-free combined image and the smooth relative phase between each shot in the multi-shot k-space data can be jointly estimated.