Accelerated two-dimensional phase-contrast for cardiovascular MRI using deep learning-based reconstruction with complex difference estimation

Purpose: To develop and validate a deep learning-based reconstruction framework for highly accelerated two-dimensional (2D) phase contrast (PC-MRI) data with accurate and precise quantitative measurements. Methods: We propose a modified DL-ESPIRiT reconstruction framework for 2D PC-MRI, comprised of an unrolled neural network architecture with a Complex Difference estimation (CD-DL). CD-DL was trained on 155 fully sampled 2D PC-MRI pediatric clinical datasets. The fully sampled data ((Formula presented.)) was retrospectively undersampled (6–11 (Formula presented.)) and reconstructed using CD-DL and a parallel imaging and compressed sensing method (PICS). Measurements of peak velocity and total flow were compared to determine the highest acceleration rate that provided accuracy and precision within (Formula presented.). Feasibility of CD-DL was demonstrated on prospectively undersampled datasets acquired in pediatric clinical patients ((Formula presented.)) and compared to traditional parallel imaging (PI) and PICS. Results: The retrospective evaluation showed that 9 (Formula presented.) accelerated 2D PC-MRI images reconstructed with CD-DL provided accuracy and precision (bias, [95 (Formula presented.) confidence intervals]) within (Formula presented.). CD-DL showed higher accuracy and precision compared to PICS for measurements of peak velocity (2.8 (Formula presented.) [(Formula presented.), 4.5] vs. 3.9 (Formula presented.) [(Formula presented.), 4.9]) and total flow (1.8 (Formula presented.) [(Formula presented.), 3.4] vs. 2.9 (Formula presented.) [(Formula presented.), 6.9]). The prospective feasibility study showed that CD-DL provided higher accuracy and precision than PICS for measurements of peak velocity and total flow. Conclusion: In a retrospective evaluation, CD-DL produced quantitative measurements of 2D PC-MRI peak velocity and total flow with (Formula presented.) error in both accuracy and precision for up to 9 (Formula presented.) acceleration. Clinical feasibility was demonstrated using a prospective clinical deployment of our 8 (Formula presented.) undersampled acquisition and CD-DL reconstruction in a cohort of pediatric patients.