Related Experiment Video
Updated: Mar 27, 2026
10:25
Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
Published on: September 25, 2019
49.7K
Tag removal in cardiac tagged MRI images using coupled dictionary learning
Summary
This study introduces a novel coupled dictionary learning method to effectively remove tagging patterns from cardiac MRI scans. This technique improves cardiac image analysis by creating clearer, de-tagged images for better segmentation and tracking.
Area of Science:
- Medical Imaging
- Biomedical Engineering
- Cardiovascular Imaging
Background:
- Tagged Magnetic Resonance Imaging (tMRI) is crucial for assessing cardiac function.
- Tagging patterns and low contrast in tMRI pose challenges for image analysis (e.g., segmentation, tracking).
- Effective removal of tagging lines is an active area of research.
Purpose of the Study:
- To propose and evaluate a novel method for removing tagging patterns from tMRI.
- To enhance the quality of cardiac MRI for improved downstream analysis.
- To offer an alternative to frequency-domain tag removal techniques.
Main Methods:
- A coupled dictionary learning (CDL) model was developed.
- The model assumes identical sparse representations for tagged and cine MRI image patches.
- Dictionaries were learned for tagged and cine MRI spaces, enabling de-tagging using sparse codes.
Main Results:
- The proposed CDL method successfully removed tagging patterns from tagged cardiac MR images.
- Experimental results demonstrated favorable performance compared to frequency-domain tag removal methods.
- The technique facilitates the creation of de-tagged cine MRI images.
Conclusions:
- The CDL model provides an effective solution for tag removal in tMRI.
- This method enhances the utility of tMRI for quantitative cardiac function assessment.
- The approach shows promise for improving cardiac image processing and analysis workflows.
