Related Experiment Video
Updated: Sep 21, 2025
12:18
Magnetic Resonance Elastography Methodology for the Evaluation of Tissue Engineered Construct Growth
Published on: February 9, 2012
12.5K
OSCILLATE: A low-rank approach for accelerated magnetic resonance elastography
Grace McIlvain1, Alexander M Cerjanic1,2, Anthony G Christodoulou3
1Department of Biomedical Engineering, University of Delaware, Newark, Delaware, USA.
Magnetic Resonance in Medicine
|June 1, 2022
Summary
Accelerated Magnetic Resonance Elastography (MRE) using OSCILLATE significantly reduces scan times. This novel low-rank reconstruction method accurately captures brain mechanical properties, enabling faster MRE acquisition.
Area of Science:
- Medical Imaging
- Biophysics
- Neuroscience
Background:
- Magnetic Resonance Elastography (MRE) characterizes in vivo brain mechanical properties.
- Current MRE acquisition is time-consuming, limiting resolution and applicability in certain patient groups.
Purpose of the Study:
- To develop an accelerated MRE acquisition method using low-rank image reconstruction.
- To exploit inherent spatiotemporal correlations in MRE data for faster scanning.
Main Methods:
- Proposed OSCILLATE (Observing Spatiotemporal Correlations for Imaging with Low-rank Leveraged Acceleration in Turbo Elastography) method.
- Alternating k-space sampling with a reduction factor (R_OSC).
- Joint low-rank image reconstruction using a temporal basis from a low-resolution navigator.
Main Results:
- MRE displacement data showed 96.1% of energy captured at rank L=12 (from full rank 24).
- Retrospective undersampling (R_OSC=2) yielded accurate stiffness maps (5.8% voxel-wise error).
- Prospective undersampling (R_OSC=2) showed minimal fidelity loss (1.0% global stiffness error).
Conclusions:
- OSCILLATE enables whole-brain MRE at 2mm isotropic resolution in under 2 minutes.
- The method accelerates MRE acquisition without compromising material property map fidelity.
