Cross correlation-based misregistration correction for super resolution T₂ -weighted spin-echo images: application to...

Area of Science:

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

Background:

• Super-resolution imaging in MRI requires precise slice alignment.
• Subtle slice-to-slice positional inconsistencies can arise during overlapped slice acquisition.
• T2-weighted spin-echo multislice imaging is susceptible to these motion artifacts.

Purpose of the Study:

• To develop a retrospective correction method for subtle slice-to-slice positional inconsistencies.
• To enhance super-resolution imaging in T2-weighted spin-echo multislice MRI.
• To improve image registration accuracy for overlapped slices.

Main Methods:

• Utilized cross-correlation to measure slice-to-slice displacement in overlapped MRI slices.
• Applied Z-dependent filtering to preserve true object shape during correction.
• Tested the method in a controlled phantom study and in vivo on 16 prostate MRI patients.

Main Results:

• The correlation method achieved high accuracy (within 0.03 mm/pass) and precision (±0.20 mm).
• Corrected images visually represented the true object more accurately.
• Motion-corrected axial and sagittal reformats in patient studies were significantly superior.

Conclusions:

• The developed retrospective correlation-based method significantly improves slice-to-slice registration.
• This enhancement is crucial for effective super-resolution imaging using overlapped MRI slices.
• The technique offers a valuable tool for improving MRI diagnostic accuracy.