Chemical-Shift-Encoded Imaging via Phase Unwrapping for Water-Fat Separation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing water-fat separation methods in magnetic resonance chemical-shift-encoded imaging rely on specific initial information, leading to algorithm instability and poor performance when acquisition parameters do not meet certain conditions.
Innovation Solution
A chemical-shift-encoded imaging method based on phase unwrapping that transforms the problem of selecting phasor candidates into a phase unwrapping problem, using phase conversion and unwrapping techniques to stabilize and accurately separate chemical components.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing water-fat separation methods (region-growing method based on seed points, transition region extraction method based on water-fat transition regions) are used, then chemical-shift-encoded imaging can be performed, but algorithm stability deteriorates and performance becomes poor when acquisition parameters do not meet specific conditions
Solution Approach 1:
The patent transforms the water-fat separation problem from relying on specific initial information (seed points, transition regions) to using phase unwrapping of phasor candidate solutions. This changes the fundamental parameter approach from region-based segmentation to phase-based mathematical transformation, enabling the algorithm to work stably across different acquisition parameters without requiring specific initial conditions.
Solution Approach 2:
The patent replaces the mechanical/iterative region-growing approach with a mathematical phase unwrapping system. Instead of iteratively growing regions from seed points or extracting transition regions through complex algorithms, the invention uses direct phase unwrapping mathematics to determine the correct phasor solution, eliminating the need for iterative mechanical processes that fail under certain acquisition conditions.
2Adaptability or versatility
If phase unwrapping method is used to transform the problem of selecting phasor candidates into a phase unwrapping problem, then adaptability to different acquisition parameters is improved, but computational complexity increases
Solution Approach 1:
The patent segments the complex water-fat separation problem into distinct computational stages: obtaining initial phasor candidate solutions, performing phase unwrapping on these candidates, and selecting the correct solution based on unwrapped phase continuity. This segmentation transforms an intractable single-step problem into a manageable multi-stage process, reducing overall computational complexity while maintaining adaptability.
Solution Approach 2:
The patent introduces phase unwrapping as an intermediary mathematical operation between obtaining phasor candidate solutions and determining the final water-fat separation result. This intermediary step provides a systematic way to resolve phase ambiguity without requiring complex iterative optimization, thereby managing computational complexity while achieving universal adaptability.
Data Source
AI summary
The disclosure discloses a chemical-shift-encoded imaging method and apparatus based on phase unwrapping, and a device. Comprises: performing phase conversion on the phasor candidate solution for the purpose of enabling a difference between a correct solution and an inverse decomposition solution of the phasor candidate solution to be within a set range, and on the basis of a phase unwrapping method, performing determination to obtain an intermediate phasor solution; determining a true phase of the intermediate phasor solution, and converting the true phase to a phasor candidate solution space to determine a target phasor solution; and on the basis of the target phasor solution, determining a first chemical component signal and a second chemical component signal, and on the basis of the first chemical component signal and/or the second chemical component signal, performing chemical-shift-encoded imaging.


