Related Experiment Video
Updated: Jun 10, 2025
Quantitative Magnetic Resonance Imaging of Skeletal Muscle Disease
Published on: December 18, 2016
Data Consistent Deep Rigid MRI Motion Correction
Nalini M Singh1, Neel Dey1, Malte Hoffmann2,3
1Massachusetts Institute of Technology.
Abstract:
Motion artifacts are a pervasive problem in MRI, leading to misdiagnosis or mischaracterization in population-level imaging studies. Current retrospective rigid intra-slice motion correction techniques jointly optimize estimates of the image and the motion parameters. In this paper, we use a deep network to reduce the joint image-motion parameter search to a search over rigid motion parameters alone. Our network produces a reconstruction as a function of two inputs: corrupted k-space data and motion parameters. We train the network using simulated, motion-corrupted k-space data generated with known motion parameters. At test-time, we estimate unknown motion parameters by minimizing a data consistency loss between the motion parameters, the network-based image reconstruction given those parameters, and the acquired measurements. Intra-slice motion correction experiments on simulated and realistic 2D fast spin echo brain MRI achieve high reconstruction fidelity while providing the benefits of explicit data consistency optimization. Our code is publicly available at https://www.github.com/nalinimsingh/neuroMoCo.
Insights
This study introduces a deep learning method for correcting motion artifacts in MRI scans. The novel approach enhances image reconstruction fidelity and accuracy in population imaging studies.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Neuroscience
Background:
- Motion artifacts are a significant challenge in Magnetic Resonance Imaging (MRI), potentially leading to misdiagnosis in large-scale studies.
- Existing retrospective rigid intra-slice motion correction methods involve complex joint optimization of image and motion parameters.
Purpose of the Study:
- To develop a deep learning-based method to simplify motion correction in MRI by decoupling image reconstruction from motion parameter estimation.
- To improve the accuracy and efficiency of retrospective rigid intra-slice motion correction for MRI.
Main Methods:
- A deep neural network was trained using simulated, motion-corrupted k-space data with known motion parameters.
- The network takes corrupted k-space data and motion parameters as input to generate an image reconstruction.
- At test time, unknown motion parameters are estimated by minimizing a data consistency loss.
Main Results:
- The proposed method achieves high reconstruction fidelity in simulated and realistic 2D fast spin echo brain MRI data.
- The approach successfully reduces the joint image-motion parameter search to a search over rigid motion parameters alone.
- Explicit data consistency optimization is maintained throughout the process.
Conclusions:
- The deep learning approach effectively corrects intra-slice motion artifacts in MRI, enhancing image quality and reliability.
- This method offers a more efficient and accurate solution for motion correction in population-level neuroimaging studies.
- The developed code is publicly available to facilitate further research and application.
