K-CC-MoCo: A fast respiratory motion correction in coil-compressed K-space for highly accelerated first-pass perfusion cardiac MRI

Motivation: Motion correction in free-breathing first-pass perfusion cardiac MRI (FPP-CMR) is usually done in the image domain, requiring initial reconstruction. This hinders its use in model-based and deep-learning reconstructions from highly accelerated acquisitions. Goal(s): To estimate and correct respiratory motion in free-breathing FPP-CMR directly in k-space. Approach: We propose an inter-frame rigid motion correction formulated in k-space with the normalized-cross-correlation objective function. An ROI-based coil-compression approach was employed to focus the optimization on the heart. The method was tested using a digital phantom and real free-breathing acquisitions with different accelerations. Results: The proposed approach outperforms image-based correction in acquisitions with accelerations up to 50x. Impact: The k-space-based motion correction outperforms image-based correction in free-breathing FPP-CMR acquisitions accelerated up to 50x. This method can estimate/correct respiratory motion in k-space without an initial reconstruction, thereby enabling its use for model-based and/or deep-learning reconstructions from highly accelerated scans.