Related Experiment Video
Updated: Jun 30, 2025
10:44
Three-Dimensional Phase Resolved Functional Lung Magnetic Resonance Imaging
Published on: June 21, 2024
490
Acceleration of High-Resolution 3D MR Fingerprinting via a Graph Convolutional Network
Feng Cheng1, Yong Chen2, Xiaopeng Zong3
1Department of Computer Science, University of North Carolina, Chapel Hill, NC, USA.
Medical Image Computing and Computer-Assisted Intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention
|March 20, 2024
Summary
This study introduces a deep learning method to accelerate 3D Magnetic Resonance Fingerprinting (MRF) acquisition. The new approach significantly speeds up imaging, making advanced tissue property quantification more clinically viable.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Biophysics
Background:
- Magnetic Resonance Fingerprinting (MRF) enables rapid, simultaneous quantification of multiple tissue properties.
- Current 3D MRF methods require further speed improvements for widespread clinical adoption.
Purpose of the Study:
- To develop a novel deep learning approach for accelerating 3D MRF acquisition.
- To enhance acquisition speed along the slice-encoding direction in k-space.
Main Methods:
- Introduction of a graph-based convolutional neural network tailored for non-Cartesian spiral trajectories used in MRF.
- Implementation of the deep learning model to accelerate 3D MRF data acquisition.
Main Results:
- Achieved significant acceleration in 3D MRF acquisition speed.
- Improved tissue quantification accuracy compared to existing state-of-the-art methods.
- Enabled whole-brain coverage with high spatial resolution in under 5 minutes.
Conclusions:
- The developed deep learning method makes 3D MRF more feasible for clinical applications.
- Fast and high-resolution 3D MRF imaging is now achievable within clinically relevant timeframes.
