Related Experiment Video
Updated: Oct 11, 2025
09:57
Multiplex Chemical Imaging Based on Broadband Stimulated Raman Scattering Microscopy
Published on: July 25, 2022
4.1K
High-Quality MR Fingerprinting Reconstruction Using Structured Low-Rank Matrix Completion and Subspace Projection
IEEE Transactions on Medical Imaging
|December 6, 2021
Summary
This study introduces a novel algorithm to reduce artifacts in magnetic resonance fingerprinting (MRF) imaging. The method improves the accuracy of quantitative tissue parameter maps derived from undersampled MRF data.
Area of Science:
- Medical Imaging
- Biophysics
- Signal Processing
Background:
- Magnetic Resonance Fingerprinting (MRF) offers fast multiparametric quantitative imaging.
- Artifacts from undersampled MRF data lead to inaccurate tissue parameter maps.
Purpose of the Study:
- To develop an algorithm for accurate MRF image reconstruction and parameter mapping.
- To address artifacts caused by undersampled data acquisition in MRF.
Main Methods:
- Modeling 3D MRF data as piecewise smooth signals with localized discontinuities.
- Exploiting low-rank properties of structured Toeplitz matrices from Fourier measurements.
- Employing a subspace projection scheme and an iterative two-stage algorithm for regularized problem solving.
Main Results:
- The proposed algorithm significantly improves MRF time-series image reconstruction.
- More accurate quantitative tissue parameter maps are generated compared to existing methods.
- Demonstrated effectiveness through numerical experiments.
Conclusions:
- The novel iterative two-stage algorithm effectively reduces artifacts in MRF imaging.
- This approach enhances the accuracy of quantitative parameter mapping in MRF.
- The method shows promise for improved diagnostic capabilities in quantitative MRI.
