Accelerated cardiac cine with spatio‐coil regularized deep learning reconstruction

• Published in: Magnetic resonance in medicine Vol. 93; no. 3; pp. 1132 - 1148
• Main Authors: Demirel, Omer Burak, Ghanbari, Fahime, Morales, Manuel Antonio, Pierce, Patrick, Johnson, Scott, Rodriguez, Jennifer, Street, Jordan Amy, Nezafat, Reza
• Format: Journal Article
• Language: English
• Published: United States Wiley Subscription Services, Inc 01.03.2025
• Subjects: Adult; Algorithms; cine imaging; Coils; Deep Learning; Ejection fraction; Error analysis; Female; Heart; Heart - diagnostic imaging; Heart Ventricles - diagnostic imaging; Humans; Image acquisition; Image Interpretation, Computer-Assisted - methods; Image Processing, Computer-Assisted - methods; Image reconstruction; iterative reconstruction; Magnetic Resonance Imaging, Cine - methods; Male; Regularization; Retrospective Studies; Ventricle; Ventricular Function, Left - physiology; deep learning; iterative reconstruction; cine imaging
• ISSN: 0740-3194; 1522-2594; 1522-2594
• DOI: 10.1002/mrm.30337

Purpose To develop an iterative deep learning (DL) reconstruction with spatio‐coil regularization and multichannel k‐space data consistency for accelerated cine imaging. Methods This study proposes a Spatio‐Coil Regularized DL (SCR‐DL) approach for iterative deep learning reconstruction incorporating multicoil information in data consistency and regularizer. SCR‐DL uses shift‐invariant convolutional kernels to interpolate missing k‐space lines and reconstruct individual coil images, followed by a regularizer that operates simultaneously across spatial and coil dimensions using learned image priors. At 8‐fold acceleration, SCR‐DL was compared with Generalized Autocalibrating Partially Parallel Acquisition (GRAPPA), sensitivity encoding (SENSE)‐based DL and spatio‐temporal regularized (STR)–DL reconstruction. In the retrospective undersampled cine, images were quantitatively evaluated using normalized mean square error (NMSE) and structural similarity index measure (SSIM). Additionally, agreement for left‐ventricular ejection fraction and left‐ventricular mass were assessed using prospectively accelerated cine images at 2‐fold and 8‐fold accelerations. Results The SCR‐DL algorithm successfully reconstructed highly accelerated cine images. SCR‐DL had significant improvements in NMSE (0.03 ± 0.02) and SSIM (91.4% ± 2.7%) compared with GRAPPA (NMSE: 0.09 ± 0.04, SSIM: 69.9% ± 11.1%; p < 0.001), SENSE‐DL (NMSE: 0.07 ± 0.04, SSIM: 86.9% ± 3.2%; p < 0.001), and STR‐DL (NMSE: 0.04 ± 0.03, SSIM: 90.0% ± 2.5%; p < 0.001) with retrospective undersampled cine. Despite the 3‐fold reduction in scan time, there was no difference between left‐ventricular ejection fraction (59.8 ± 4.5 vs. 60.8 ± 4.8, p = 0.46) or left‐ventricular mass (73.6 ± 19.4 g vs. 73.2 ± 19.7 g, p = 0.95) between R = 2 and R = 8 prospectively accelerated cine images. Conclusions SCR‐DL enabled highly accelerated cardiac cine imaging, significantly reducing breath‐hold time. Compared with GRAPPA or SENSE‐DL, images reconstructed with SCR‐DL showed superior NMSE and SSIM.