DEEP‐DISORDER: Motion Correction in 3D MRI via Segment Reconstruction and Registration

3D MR image acquisition is inherently time intensive, rendering it susceptible to patient motion during scanning. This may introduce significant blurring and artifacts, potentially necessitating reacquisition. We propose a modular framework to retrospectively correct for intrascan motion in 3D brain MRI, without active motion tracking. Serving as the backbone of our approach is an existing distributed and incoherent sampling scheme (DISORDER), combined with a fast network trained for highly undersampled reconstruction. This enables approximate reconstructions of anatomy after every few seconds, using only a tiny fraction of k-space data (0. 942 ± 0. 026 0. 942 0. 026 to 0. 992 ± 0. 003 0. 992 0. 003 SSIM for the retrospective scans. The prospective scans improved from 0. 915 ± 0. 024 0. 915 0. 024 to 0. 936 ± 0. 014 0. 936 0. 014 SSIM after correction in the case of gradual motion and from 0. 764 ± 0. 008 0. 764 0. 008 to 0. 923 ± 0. 011 0. 923 0. 011 SSIM for extreme motion. In conclusion, the proposed approach, that is free of external tracking devices or navigators, successfully estimated and corrected 3D motion between small subportions of a scan. This resulted in vastly improved image quality, making volumetric MRI substantially more tolerant to motion.