Real-time assessment of right and left ventricular volumes and function in children using high spatiotemporal resolution spiral bSSFP with compressed sensing

Abstract

BackgroundReal-time cardiovascular magnetic resonance (CMR) assessment of ventricular volumes and function enables data acquisition during free-breathing. The requirement for high spatiotemporal resolution in children necessitates the use of highly accelerated imaging techniques.MethodsA novel real-time balanced steady state free precession (bSSFP) spiral sequence reconstructed using Compressed Sensing (CS) was prospectively validated against the breath-hold clinical standard for assessment of ventricular volumes in 60 children with congenital heart disease. Qualitative image scoring, quantitative image quality, as well as evaluation of biventricular volumes was performed. Standard BH and real-time measures were compared using the paired t-test and agreement for volumetric measures were evaluated using Bland Altman analysis.ResultsAcquisition time for the entire short axis stack (~ 13 slices) using the spiral real-time technique was ~ 20 s, compared to ~ 348 s for the standard breath hold technique. Qualitative scores reflected more residual aliasing artefact (p < 0.001) and lower edge definition (p < 0.001) in spiral real-time images than standard breath hold images, with lower quantitative edge sharpness and estimates of image contrast (p < 0.001).There was a small but statistically significant (p < 0.05) overestimation of left ventricular (LV) end-systolic volume (1.0 ± 3.5 mL), and underestimation of LV end-diastolic volume (− 1.7 ± 4.6 mL), LV stroke volume (− 2.6 ± 4.8 mL) and LV ejection fraction (− 1.5 ± 3.0%) using the real-time technique. We also observed a small underestimation of right ventricular stroke volume (− 1.8 ± 4.9 mL) and ejection fraction (− 1.4 ± 3.7%) using the real-time imaging technique. No difference in inter-observer or intra-observer variability were observed between the BH and real-time sequences.ConclusionsReal-time bSSFP imaging using spiral trajectories combined with a compressed sensing reconstruction showed good agreement for quantification of biventricular metrics in children with heart disease, despite slightly lower image quality. This technique holds the potential for free breathing data acquisition, with significantly shorter scan times in children.Electronic supplementary materialThe online version of this article (10.1186/s12968-018-0500-9) contains supplementary material, which is available to authorized users.

Perturbed spiral real‐time phase‐contrast MR with compressive sensing reconstruction for assessment of flow in children

“…In the clinical study, both the GAS perturbed and GAS uniform acquisitions produced good-quality magnitude images with little residual artifacts. This is in keeping with previous studies that combined spiral imaging with CS reconstruction 12 and should be expected, as regularization was optimized for image quality. However, velocity and flow curves extracted from GAS uniform data were blurred in time compared with the reference standard Cartesian-gated data.…”

“…Spiral trajectories are of particular interest for PCMR due to short TEs and highly efficient filling of k‐space. Golden‐angle spiral imaging with CS reconstruction has been shown to be sufficient for real‐time cine data . However, we have shown in a pilot study that CS reconstruction can cause temporal blurring of PCMR data that could result in underestimation of clinically important metrics.…”

Deblurring for spiral real‐time MRI using convolutional neural networks

“…5,6 Spiral imaging is also widely used for real-time MRI (RT-MRI), for which the capability of capturing rapid motion is crucial, such as in cardiac imaging and speech production imaging. [6][7][8][9][10][11][12] One major limitation of spiral sampling is image blurring due to off-resonance, 13 which causes the accumulation of phase error along the readout in k-space domain, resulting in blurring and/or signal loss in the image domain. To date, this artifact remains the predominant challenge for several RT-MRI applications: In speech RT-MRI, it degrades image quality primarily at the air-tissue boundaries, which include the vocal tract articulators of interest.…”

Revival and modification of the Mustard operation

“…The 3 atrial functions can be measured noninvasively by quantifying phasic changes in atrial area or volume using acoustic border detection, vector velocity mapping using 2-dimensional echocardiography, 30 speckle tracking, 31 or CMR. 32,33 In the current study, we used CMR to define atrial functions after the Mustard operation and report the differences between the classic and modified techniques in a representative sample of 21 patients. These preliminary results show enhanced reservoir, conduit, and contractile atrial functions after the modified technique when compared with the classic technique.…”