Retrospective motion correction through multi-average k-space data elimination (REMAKE) for free-breathing cardiac...

Area of Science:

• Medical Imaging
• Cardiovascular MRI
• Image Reconstruction

Background:

• Free-breathing cine MRI with multiple averages is crucial for cardiac assessment.
• Motion artifacts degrade image quality and diagnostic accuracy in cardiac MRI.
• Existing reconstruction methods struggle to effectively correct for motion in free-breathing scans.

Purpose of the Study:

• To develop and validate motion-robust reconstruction techniques for free-breathing cine MRI.
• To improve image quality and diagnostic yield in cardiac MRI by reducing motion artifacts.
• To introduce REMAKE and REMAKE+ as advanced methods for retrospective motion correction.

Main Methods:

• Developed REMAKE (Retrospective motion correction through multiple average k-space data elimination) by iteratively removing motion-corrupted k-space segments.
• Introduced REMAKE+ to mitigate SNR loss by averaging multiple reconstructions with different initial conditions.
• Validated REMAKE and REMAKE+ against standard signal averaging reconstruction in phantom studies and 15 patients.

Main Results:

• REMAKE/REMAKE+ significantly reduced motion artifacts in phantom studies (RMSE reduction).
• In patients, REMAKE/REMAKE+ improved myocardial sharpness, overall image quality, and the rate of diagnostic quality images compared to standard methods.
• While standard averaging showed higher initial SNR/CNR, REMAKE+ demonstrated comparable or improved SNR/CNR with superior motion correction.

Conclusions:

• REMAKE and REMAKE+ offer superior motion correction for free-breathing cine MRI.
• These techniques enhance myocardial sharpness and improve the diagnostic quality of cardiac MRI scans.
• REMAKE/REMAKE+ represent advancements in MRI reconstruction for improved cardiovascular imaging.