Revised motion estimation algorithm for PROPELLER MRI.

Area of Science:

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

Background:

• Motion artifacts are a significant challenge in Magnetic Resonance Imaging (MRI), particularly in fast imaging techniques like PROPELLER.
• Accurate estimation of motion-induced data shifts is crucial for effective motion correction in MRI.
• Existing algorithms may be sensitive to noise and do not fully leverage correlations between different data segments.

Purpose of the Study:

• To introduce and evaluate a novel algorithm for estimating data shifts (rotation and translation) in motion-corrected PROPELLER MRI.
• The new algorithm jointly estimates shifts for all blades, prioritizing strong and noise-robust blade-pair correlations.

Main Methods:

• PROPELLER MRI scans were acquired from three volunteers exhibiting varying degrees of head motion.
• Data were reconstructed using both the original and the new motion estimation algorithms.
• Two radiologists independently assessed 216 image pairs, comparing the quality of images reconstructed with each method.

Main Results:

• Across 432 total scores, the new algorithm was rated substantially better than the original method 11 times.
• The new algorithm was never rated substantially worse than the original method.
• These results indicate a favorable comparison for the new algorithm's motion estimation capabilities.

Conclusions:

• The novel algorithm demonstrates a favorable performance in estimating bulk motion for PROPELLER MRI compared to the existing method.
• The study highlights the potential of joint estimation and robust correlations for improved motion correction.
• Future work will involve a larger patient study to further validate the algorithm's efficacy.