Related Experiment Video
Updated: Jul 1, 2025
09:55
Author Spotlight: Using Hyperpolarized Xenon-129 MRI to Study Lung Diseases
Published on: January 5, 2024
1.2K
Sampling pattern discrepancy in the application of compressed sensing hyperpolarized xenon-129 lung MRI.
Mitra Tavakkoli1,2, Sarah Svenningsen3,4,5, Yonni Friedlander3
1Imaging Research Centre, St. Joseph's Healthcare Hamilton, Hamilton, Ontario, Canada.
NMR in Biomedicine
|February 29, 2024
Summary
Random undersampling in hyperpolarized 129Xe MRI ventilation imaging creates significant image quality variations, especially below 60% sampling. Optimizing sampling patterns is crucial for compressed sensing to be effective in highly undersampled 129Xe lung datasets.
Area of Science:
- Medical Imaging
- Pulmonology
- Magnetic Resonance Imaging
Background:
- Hyperpolarized (HP) 129Xe ventilation MRI allows breath-hold imaging but is challenging for patients.
- Compressed Sensing (CS) accelerates MRI acquisition but undersampled images vary.
- Image quality metrics like SNR and SSIM are sensitive to undersampling patterns.
Purpose of the Study:
- To investigate the impact of random undersampling on HP 129Xe lung ventilation MRI quality.
- To determine the minimum sampling ratio required for reliable CS reconstruction.
- To highlight the importance of optimized sampling patterns for CS in 129Xe MRI.
Main Methods:
- Acquired fully sampled 2D multi-slice HP 129Xe lung ventilation images in 20 subjects (10 healthy, 10 asthma).
- Generated 500 random undersampling patterns at ratios from 10% to 80%.
- Reconstructed images using Parallel Imaging and Compressed Sensing (PICS) and compared SNR, SSIM, and sidelobe to peak ratio.
Main Results:
- Significant variation in SNR and SSIM was observed across different random undersampling patterns at all ratios.
- Image quality metric variation increased with higher undersampling.
- Random undersampling below 60% sampling ratio led to substantial variability, despite meeting CS incoherency criteria.
Conclusions:
- Random undersampling poses a significant challenge for HP 129Xe MRI at low sampling ratios (<60%).
- The variability in image quality necessitates careful consideration of sampling patterns.
- Optimizing the sampling pattern is essential for maximizing the potential of CS in highly undersampled 129Xe lung MRI.
