Pediatric Track Session
SCMR 22nd Annual Scientific Sessions
Eric Schrauben, PhD
Research Fellow
Hospital for Sick Children
Jessie Mei Lim, BSc
PhD Student
University of Toronto
Datta Goolaub
PhD Student
The Hospital for Sick Children
Davide Marini, MD, PhD
Staff Cardiology - Assistant Professor
The Sickkids Hospital
Mike Seed, MD
Staff Cardiologist and Radiologist
SickKids
Christopher Macgowan, PhD
Senior Scientist
The Hospital for Sick Children
Background:
Neonatal cardiovascular MR exams are technically challenging because of the need for high spatial resolution and the presence of bulk and respiratory motions. These sources of motion are more pronounced in neonates with complex congenital heart disease (CHD), which are preferably scanned in a feed-and-sleep manner to avoid sedation. Probing of complex 3D blood flow patterns (for example, with 4D flow MRI [1]) in this population is therefore particularly difficult.
3D radial trajectories offer an alternative to conventional Cartesian studies, providing volumetric coverage with high isotropic resolution in a short scan time by allowing undersampling at the cost of incoherent streaking artifacts [2]. Furthermore, by repeatedly sampling the k-space center, such radial data enable the quantification and compensation of bulk and respiratory motions.
The purpose of this study was the development of a framework for measuring and correcting motion using 3D radial MR with multi-dimensional flow sensitivity, and to implement this in neonates and young infants with CHD to investigate complex 3D blood flow patterns.
Methods: As part of an ongoing research study, 6 patients with CHD (Table 1) were imaged on a 1.5T scanner (AvantoFIT, Siemens). Acquisitions consisted of a free-running double golden angle 3D radial sequence with phase contrast velocity encoding applied along each spatial dimension [3] (Table 1). Note the high isotropic spatial resolutions (0.60 – 0.86 mm) and relatively short scan times (3.25 – 5.80 min), critical for this population. Figure 1 outlines the acquisition and reconstruction pipeline, which includes k-space based bulk motion detection and correction [4], respiratory phase detection, and compressed sensing reconstruction using spatial total variation. Due to high undersampling from short scan times, data are cardiac time-averaged for final reconstruction. Images were processed using 4D Flow v2.4 (Siemens). Quantified flows were validated using 2D PC MRI (multiple averages without respiratory gating) in individual great vessels.
Results: Acquisition, reconstruction, and processing were successfully completed in all subjects. The single pre-operative patient is shown in Figure 2A, with example segmentation, whole volume streamlines, and velocity vector visualization in the parallel great arteries. Figure 2B shows anterior positioning of the distal and branch pulmonary arteries following the LeCompte maneuver in Patient 3. Across measured flows in twenty individual vessels between both techniques, regression displayed good correlation for 3D radial flow in expiration with 2D PC MRI (slope = 1.04, R2 = 0.96, p < 0.0001).
Conclusion: This work demonstrates the feasibility of a fast, motion-robust volumetric velocity imaging technique performed in neonates and infants with CHD. Future work will extend this technique to acquire cardiac CINE imaging for dynamic visualization and assessment of these small structures with complex flow patterns.