Motion robust magnetic resonance imaging via efficient Fourier aggregation

Area of Science:

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

Background:

• Motion artifacts significantly degrade the quality of anatomical MRI scans, often rendering them unusable for clinical diagnosis.
• Existing methods for motion artifact correction in MRI have limitations, particularly when applied to reconstructed image data.

Purpose of the Study:

• To introduce and validate MOTOR-MRI, a novel technique for suppressing motion artifacts in anatomical MRI.
• To demonstrate the ability of MOTOR-MRI to recover and salvage severely corrupted MRI images.

Main Methods:

• MOTOR-MRI operates on reconstructed images, not k-space data.
• The technique involves segmenting standard MRI acquisition protocols into shorter segments.
• It efficiently aggregates locally sharp and consistent information in Fourier space to produce a motion-mitigated image.

Main Results:

• Significant qualitative and quantitative improvements in image quality were observed on T2-weighted turbo spin echo brain scans.
• The method proved effective on data acquired from both 3T and 7T scanners, even with severe motion corruption.
• MOTOR-MRI successfully recovered images that were otherwise unusable due to motion artifacts and blur.

Conclusions:

• MOTOR-MRI offers a robust solution for motion artifact suppression in anatomical MRI.
• The technique enhances image quality and can salvage corrupted scans, improving diagnostic utility.
• MOTOR-MRI can be used independently or combined with other motion correction methods.