POCS-based reconstruction of multiplexed sensitivity encoded MRI (POCSMUSE): A general algorithm for reducing...

Area of Science:

• Magnetic Resonance Imaging (MRI)
• Image Reconstruction
• Medical Imaging Signal Processing

Background:

• Iterative projection onto convex sets (POCS) reconstructs MRI images using coil sensitivity profiles as constraints.
• This POCS method is versatile, applicable to diverse pulse sequences and k-space trajectories.
• Additional constraints enhance reconstruction stability for ill-conditioned matrices.

Purpose of the Study:

• Develop a projection onto convex sets reconstruction of multiplexed sensitivity encoded MRI (POCSMUSE) algorithm.
• Reduce motion-related artifacts, specifically respiration artifacts in abdominal imaging and aliasing in diffusion-weighted imaging.
• Enhance overall image quality in challenging MRI scenarios.

Main Methods:

• Applied POCSMUSE to abdominal fast spin-echo imaging data for respiratory artifact evaluation.
• Utilized POCSMUSE to mitigate aliasing artifacts in interleaved diffusion-weighted imaging data.
• Assessed effectiveness across varying k-space trajectories and matrix condition numbers.

Main Results:

• POCSMUSE demonstrated significant reduction of motion-related artifacts.
• Effectiveness confirmed across different MRI pulse sequences, k-space trajectories, and contrasts.
• Successful application in both abdominal imaging and diffusion-weighted imaging.

Conclusions:

• POCSMUSE is a versatile post-processing algorithm for motion artifact reduction in MRI.
• The method is compatible with various pulse sequences and imaging contrasts.
• POCSMUSE can further refine images already processed by existing artifact reduction techniques.