Related Experiment Video
Updated: Dec 15, 2025
10:44
Three-Dimensional Phase Resolved Functional Lung Magnetic Resonance Imaging
Published on: June 21, 2024
957
Motion correction and noise removing in lung diffusion-weighted MRI using low-rank decomposition
Xinhui Wang1, Houjin Chen2, Qi Wan3
1School of Electronic and Information Engineering, Beijing Jiaotong University, Beijing, China.
Medical & Biological Engineering & Computing
|July 13, 2020
Summary
A new low-rank decomposition method significantly reduces blurring in lung diffusion-weighted MRI (DWI) images. This technique improves image clarity and enhances lung cancer diagnosis accuracy, offering a promising tool for medical imaging.
Area of Science:
- Medical Imaging
- Magnetic Resonance Imaging
- Image Processing
Background:
- Lung diffusion-weighted magnetic resonance imaging (DWI) is valuable for lung lesion detection and diagnosis.
- Image blurring caused by motion and physiological factors can impair diagnostic performance.
Purpose of the Study:
- To reduce blurring in lung DWI.
- To assess the impact of deblurring on lung cancer diagnosis.
Main Methods:
- A retrospective study involving 71 patients.
- Development of a motion correction and noise removal method using low-rank decomposition.
- Evaluation through qualitative and quantitative assessments and diagnostic performance metrics (AUC).
Main Results:
- Qualitative improvements include reduced lung mass deformation and alleviated tumor edge blurring.
- Quantitative analysis showed significant increases in Mutual Information (MI) and Pearson correlation coefficient (Pearson-Coff) post-decomposition.
- Area Under Curve (AUC) for lung cancer diagnosis improved from 0.731 to 0.841.
Conclusions:
- Low-rank matrix decomposition effectively reduces noise and artifacts in lung DWI.
- The method shows promise for enhancing diagnostic accuracy in lung cancer detection.
- Further research is needed to explore the full potential of this technique in lung DWI.
