Motion-robust reconstruction of multishot diffusion-weighted images without phase estimation through locally low-rank...

Area of Science:

• Medical Imaging
• Magnetic Resonance Imaging
• Image Reconstruction

Background:

• Diffusion-weighted MRI (DW-MRI) is crucial for assessing tissue microstructure.
• Multishot DW-MRI sequences are susceptible to shot-to-shot phase mismatches caused by motion.
• Existing methods often require complex phase estimation, limiting their applicability.

Purpose of the Study:

• To develop a motion-robust reconstruction method for DW-MRI.
• To resolve shot-to-shot phase mismatches without explicit phase estimation.
• To improve image quality in the presence of motion artifacts.

Main Methods:

• Proposed a novel reconstruction technique termed the shot-locally low-rank method.
• Formed spatial-shot matrices using local pixel groups from different shots.
• Applied a convex model with a locally low-rank constraint to these matrices.

Main Results:

• Demonstrated significant benefits in motion-robustness, particularly for breast imaging.
• Achieved improved image quality in vivo for brain and breast experiments.
• The reconstructed images enabled reliable phase estimation for further enhancement.

Conclusions:

• Introduced a motion-robust reconstruction method for multishot DW-MRI without explicit phase estimation.
• The shot-locally low-rank method offers advantages for neuroimaging and body imaging.
• The technique enhances image quality and can be integrated with phase-estimation-based methods.