Multi-Echo Stack-of-Stars MRI Thermometry with Pseudo-Golden Angle Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Magnetic Resonance Imaging (MRI) techniques for monitoring focused ultrasound (FUS) treatments, especially in 2D MRgFUS, face limitations such as limited field of view, partial volume effects, and motion artifacts, which hinder accurate temperature measurement and energy deposition monitoring during treatments like uterine fibroid and cancer therapy.
Innovation Solution
A 3D multi-echo stack-of-stars radial acquisition method with pseudo golden angle (PGA) sampling and k-space weighted image contrast (KWIC) temporal weighting is employed, allowing for simultaneous measurement of proton resonance frequency (PRF) shift temperature, initial signal magnitude (M(0)), and T2* values, while correcting respiration artifacts and providing high spatial and temporal resolution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If 3D MR thermometry is used to overcome field of view and coverage limitations, then spatial coverage and measurement accuracy are improved, but acquisition time increases making it clinically non-viable
Solution Approach 1:
The 3D k-space is segmented into multiple radial stacks acquired at different angles, allowing comprehensive 3D coverage to be divided into manageable 2D acquisitions that can be performed rapidly and interleaved with treatment delivery
Solution Approach 2:
Multiple radial stacks are acquired periodically at different angular orientations throughout the treatment, enabling continuous monitoring with temporal resolution matched to the periodic treatment cycles while maintaining 3D coverage
2Productivity
If 2D slices are used for monitoring, then acquisition speed is improved, but field of view and coverage are limited missing heating near skull surface and gaps between slices
Solution Approach 1:
The methodology transitions from 2D slice acquisition to 3D volumetric coverage by acquiring multiple radial stacks at different angles, adding the angular dimension to achieve comprehensive 3D monitoring while maintaining rapid acquisition speeds
3Area of stationary object
If multiple 2D slices are used to increase field of view, then coverage is improved, but gaps between slices and partial volume effects cause measurement errors
Solution Approach 1:
The 3D volume is segmented into multiple radial stacks that are angularly separated, with each stack providing overlapping coverage that eliminates gaps while maintaining sufficient spatial resolution to avoid partial volume effects
Solution Approach 2:
The radial stack acquisition geometry is specifically designed beforehand to ensure overlapping coverage regions between adjacent stacks, guaranteeing continuous 3D coverage without gaps before treatment begins
4Productivity
If temporal resolution is increased by undersampling methods, then acquisition speed is improved, but image quality and signal to noise ratio deteriorate
Solution Approach 1:
The radial stack acquisition continuously samples through the center of k-space every TR period, maintaining continuous updating of low spatial frequency information that provides robust temporal resolution without significant loss of signal quality
Solution Approach 2:
The acquired data from multiple radial stacks is combined using compressed sensing and iterative reconstruction algorithms that incorporate physiological models and constraints, providing feedback-driven enhancement that recovers signal quality from undersampled data
Data Source
AI summary
A method for producing an image of a subject using a magnetic resonance imaging (MRI) system includes acquiring a series of echo signals by sampling k-space along radial lines that each pass through the center of k-space. Each projection of the radial lines is divided into multiple echoes and successive projections are spaced by a predetermined angular distance. The series of echo signals are reconstructed into a plurality of images, wherein each image corresponds to a distinct echo signal.


