Related Experiment Videos
Retrospective intra-scan motion correction.
Marc E Bourgeois1, Frank T A W Wajer, Muriel Roth
1Laboratoire de RMN, CNRS UMR 5012, Université Claude Bernard LYON I, CPE, Domaine scientifique de la Doua, 3 rue Victor Grignard, F-69616 Villeurbanne, France.
Journal of Magnetic Resonance (San Diego, Calif. : 1997)
|August 14, 2003
Summary
This study introduces an automatic retrospective method to correct intra-scan motion in MRI by analyzing k-space data. The technique accurately estimates and corrects motion, significantly reducing artifacts in reconstructed images.
Area of Science:
- Medical Imaging
- Biophysics
- Computational Neuroscience
Background:
- Intra-scan motion during Magnetic Resonance Imaging (MRI) can degrade image quality and introduce artifacts.
- Existing motion correction methods often require external tracking or are applied post-acquisition, limiting their effectiveness for rapid, subtle movements.
- Correction directly in k-space offers a promising avenue for robust motion artifact mitigation.
Purpose of the Study:
- To develop and validate an automatic retrospective method for correcting intra-scan motion artifacts in 2D Cartesian MRI acquisitions.
- To demonstrate the feasibility of performing motion correction directly in k-space.
- To assess the accuracy and performance of the proposed method across various signal-to-noise ratios.
Main Methods:
- A novel automatic retrospective method for intra-scan motion correction in k-space.
- Utilizes a reference k-space acquisition to estimate motion parameters (rotation and translation) for each data set.
- Employs a Bayesian image reconstruction estimator capable of handling sparse, arbitrary k-space sampling.
Main Results:
- Accurate estimation of motion parameters: better than 0.1 degrees for rotation and sub-pixel accuracy for translation.
- Significant reduction of intra-scan rotation artifacts to the noise level.
- Successful reconstruction of experimental data corrupted by head motion, demonstrating clinical relevance.
Conclusions:
- The developed k-space based retrospective motion correction method is effective for 2D Cartesian MRI.
- The technique offers high accuracy in motion parameter estimation and artifact reduction.
- This approach holds potential for improving the quality and reliability of MRI scans affected by patient motion.