MRI artifact correction using sparse + low-rank decomposition of annihilating filter-based hankel matrix

Area of Science:

• Medical Imaging
• Image Processing
• Biomedical Engineering

Background:

• Magnetic Resonance Imaging (MRI) artifacts arise from system instability, patient motion, and field inhomogeneities.
• Current methods for MRI artifact correction often require additional scans, increasing time and cost.
• Artifacts are typically considered irreversible, necessitating complex workarounds.

Purpose of the Study:

• To propose a novel compressed sensing-based approach for the removal of various MRI artifacts.
• To develop a method that overcomes the limitations of existing artifact correction techniques.
• To provide a robust solution for improving the quality of MRI scans.

Main Methods:

• Utilized a sparse + low-rank decomposition framework employing a Hankel matrix.
• Leveraged the annihilating filter based low-rank Hankel matrix approach.
• Employed the alternating direction method of multipliers algorithm for cost function minimization.

Main Results:

• The proposed algorithm successfully corrected common MRI artifacts, including herringbone (crisscross), motion, and zipper artifacts.
• Experimental results demonstrated artifact correction without introducing image distortion.
• The method proved effective in identifying and removing sparse outliers representing artifacts.

Conclusions:

• The developed method offers a robust solution for correcting diverse MRI artifacts.
• The approach is particularly effective for artifacts representable as sparse outliers in k-space.
• This technique enhances the reliability and diagnostic value of MRI scans.