Publications

PURPOSE: To develop a free-breathing cardiac MR perfusion sequence with slice tracking for use after physical exercise. METHODS: We propose to use a leading navigator, placed immediately before each 2D slice acquisition, for tracking the respiratory motion and updating the slice location in real-time. The proposed sequence was used to acquire CMR perfusion datasets in 12 healthy adult subjects and 8 patients. Images were compared with the conventional perfusion (i.e., without slice tracking) results from the same subjects. The location and geometry of the myocardium were quantitatively analyzed, and the perfusion signal curves were calculated from both sequences to show the efficacy of the proposed sequence. RESULTS: The proposed sequence was significantly better compared with the conventional perfusion sequence in terms of qualitative image scores. Changes in the myocardial location and geometry decreased by 50% in the slice tracking sequence. Furthermore, the proposed sequence had signal curves that are smoother and less noisy. CONCLUSION: The proposed sequence significantly reduces the effect of the respiratory motion on the image acquisition in both rest and stress perfusion scans.

PURPOSE: To develop and evaluate an image reconstruction technique for cardiac MRI (CMR) perfusion that uses localized spatio-temporal constraints. METHODS: CMR perfusion plays an important role in detecting myocardial ischemia in patients with coronary artery disease. Breath-hold k-t-based image acceleration techniques are typically used in CMR perfusion for superior spatial/temporal resolution and improved coverage. In this study, we propose a novel compressed sensing-based image reconstruction technique for CMR perfusion, with applicability to free-breathing examinations. This technique uses local spatio-temporal constraints by regularizing image patches across a small number of dynamics. The technique was compared with conventional dynamic-by-dynamic reconstruction, and sparsity regularization using a temporal principal-component (pc) basis, as well as zero-filled data in multislice two-dimensional (2D) and three-dimensional (3D) CMR perfusion. Qualitative image scores were used (1 = poor, 4 = excellent) to evaluate the technique in 3D perfusion in 10 patients and five healthy subjects. On four healthy subjects, the proposed technique was also compared with a breath-hold multislice 2D acquisition with parallel imaging in terms of signal intensity curves. RESULTS: The proposed technique produced images that were superior in terms of spatial and temporal blurring compared with the other techniques, even in free-breathing datasets. The image scores indicated a significant improvement compared with other techniques in 3D perfusion (x-pc regularization, 2.8 ± 0.5 versus 2.3 ± 0.5; dynamic-by-dynamic, 1.7 ± 0.5; zero-filled, 1.1 ± 0.2). Signal intensity curves indicate similar dynamics of uptake between the proposed method with 3D acquisition and the breath-hold multislice 2D acquisition with parallel imaging. CONCLUSION: The proposed reconstruction uses sparsity regularization based on localized information in both spatial and temporal domains for highly accelerated CMR perfusion with potential use in free-breathing 3D acquisitions.

PURPOSE: To investigate the efficacy of a novel respiratory motion scheme, where only the center of k-space is gated using respiratory navigators, versus a fully respiratory-gated acquisition for three-dimensional flow imaging. METHODS: Three-dimensional flow images were acquired axially using a gradient echo sequence in a volume, covering the ascending and descending aorta, and the pulmonary artery bifurcation in 12 healthy subjects (33.2 ± 15.8 years; five men). For respiratory motion compensation, two gating and tracking strategies were used with a 7-mm gating window: (1) All of k-space acquired within the gating window (fully gated) and (2) central k-space acquired within the gating window, and the remainder of k-space acquired without any gating (center gated). Each scan was repeated twice. Stroke volume, mean flow, peak velocity, and signal-to-noise-ratio measurements were performed both on the ascending and on the descending aorta for all acquisitions, which were compared using a linear mixed-effects model and Bland-Altman analysis. RESULTS: There were no statistical differences between the fully gated and the center-gated strategies for the quantification of stroke volume, peak velocity, and mean flow, as well as the signal-to-noise-ratio measurements. Furthermore, the proposed center-gated strategy had significantly shorter acquisition time compared to the fully gated strategy (13:19 ± 3:02 vs. 19:35 ± 5:02, P < 0.001). CONCLUSIONS: The proposed novel center-gated strategy for three-dimensional flow MRI allows for markedly shorter acquisition time without any systematic variation in quantitative flow measurements in this small group of healthy volunteers.

PURPOSE: Coronary magnetic resonance angiography (MRA) is commonly performed with diaphragmatic navigator (NAV) gating to compensate for respiratory motion, but this approach is inefficient as data must be reacquired when it is outside the acceptance window. We therefore developed and validated a motion compensation technique based on three-dimensional (3D) spatial registration in which data are accepted throughout the respiratory cycle. METHODS: A novel respiratory motion compensation method was implemented that acquires a low-resolution 3D-image of the heart (3D-LOC) just prior to coronary MRA data acquisition. 3D-LOC volumes were registered to the first 3D-LOC to estimate the respiratory-induced heart motion and to modify the coronary MRA data and reconstruct motion-corrected images. Whole-heart coronary MRA datasets were acquired from nine healthy subjects using a diaphragmatic NAV and using 3D-LOC. RESULTS: There was no significant difference between the subjective image score of NAV and 3D-LOC in three main coronary branches. The vessel sharpness of 3D-LOC was higher than NAV in the right (0.44 ± 0.08 vs. 0.49 ± 0.08; P = 0.055) and left circumflex arteries (0.49 ± 0.05 vs. 0.52 ± 0.04; P = 0.039). Scan time for 3D-LOC was significantly shorter than NAV (4.3 ± 0.6 vs. 8.3 ± 2.3 min; P = 0.004). CONCLUSION: Compared to NAV gating, 3D-LOC for coronary MRA reduces scan time by nearly 50% without compromising image quality.

PURPOSE: To improve compressed sensing (CS) reconstruction of accelerated breath-hold (BH) radial cine magnetic resonance imaging (MRI) by exploiting auxiliary data acquired between different BHs. MATERIALS AND METHODS: Cardiac function is usually assessed using segmented cine acquisitions over multiple BHs to cover the entire left ventricle (LV). Subjects are given a resting period between adjacent BHs, when conventionally no data are acquired and subjects rest in the scanner. In this study the resting periods between BHs were used to acquire additional free-breathing (FB) data, which are subsequently used to generate a sparsity constraint for each cardiac phase. Images reconstructed using the proposed sparsity constraint were compared with conventional CS using a composite image generated by averaging different cardiac phases. The efficacy of the proposed reconstruction was compared using indices of LV function and blood-myocardium sharpness. RESULTS: The proposed method provided accurate LV ejection fraction measurements for 33% and 20% sampled datasets compared with fully sampled reference images, and showed 14% and 11% higher blood-myocardium border sharpness scores compared to the conventional CS. CONCLUSION: The FB data acquired during resting periods can be efficiently used to improve the image quality of the undersampled BH data without increasing the total scan time.

PURPOSE: To develop arrhythmia-insensitive inversion recovery sequences for improved visualization of myocardial scar and quantification of diffuse fibrosis. METHODS: A novel preparation pre-pulse, called saturation pulse prepared heart-rate-independent inversion recovery, is introduced, which consists of a combination of saturation and inversion pulses to remove the magnetization history in each heartbeat in late gadolinium enhancement (LGE) imaging and eliminate the need for rest periods in T1 mapping. The proposed LGE and T1 mapping sequences were evaluated against conventional LGE and modified Look-Locker inversion sequences using numerical simulations, phantom and imaging in healthy subjects and patients with suspected or known cardiovascular disease. RESULTS: Simulations and phantom experiments show that the saturation pulse prepared heart-rate-independent inversion recovery pre-pulse in LGE reduces ghosting artifacts and results in perfect nulling of the healthy myocardium in the presence of arrhythmia. In T1 mapping, saturation pulse prepared heart-rate-independent inversion recovery results in (a) reduced scan time (17 vs. 9 heartbeats), (b) insensitivity to heart rate for long T1, and (c) increased signal homogeneity for short T1. LGE images in a patient in atrial fibrillation during the scan show improved myocardial nulling. In vivo T1 maps demonstrate increased signal homogeneity in blood pools and myocardium. CONCLUSION: The proposed sequences are insensitive to heart rate variability, yield improved LGE images in the presence of arrhythmias, as well as T1 mapping with shorter scan times.

BACKGROUND: To investigate the feasibility of accelerated electrocardiogram (ECG)-triggered contrast enhanced pulmonary vein magnetic resonance angiography (CE-PV MRA) with isotropic spatial resolution using compressed sensing (CS). METHODS: Nineteen patients (59±13 y, 11 M) referred for MR were scanned using the proposed accelerated free breathing ECG-triggered 3D CE-PV MRA sequence (FOV=340×340×110 mm3, spatial resolution=1.5×1.5×1.5 mm3, acquisition window=140 ms at mid diastole and CS acceleration factor=5) and a conventional first-pass breath-hold non ECG-triggered 3D CE-PV MRA sequence. CS data were reconstructed offline using low-dimensional-structure self-learning and thresholding reconstruction (LOST) CS reconstruction. Quantitative analysis of PV sharpness and subjective qualitative analysis of overall image quality were performed using a 4-point scale (1: poor; 4: excellent). RESULTS: Quantitative PV sharpness was increased using the proposed approach (0.73±0.09 vs. 0.51±0.07 for the conventional CE-PV MRA protocol, p<0.001). There were no significant differences in the subjective image quality scores between the techniques (3.32±0.94 vs. 3.53±0.77 using the proposed technique). CONCLUSIONS: CS-accelerated free-breathing ECG-triggered CE-PV MRA allows evaluation of PV anatomy with improved sharpness compared to conventional non-ECG gated first-pass CE-PV MRA. This technique may be a valuable alternative for patients in which the first pass CE-PV MRA fails due to inaccurate first pass timing or inability of the patient to perform a 20-25 seconds breath-hold.

BACKGROUND: Non-Cartesian trajectories are used in a variety of fast imaging applications, due to the incoherent image domain artifacts they create when undersampled. While the gridding technique is commonly utilized for reconstruction, the incoherent artifacts may be further removed using compressed sensing (CS). CS reconstruction is typically done using conjugate-gradient (CG) type algorithms, which require gridding and regridding to be performed at every iteration. This leads to a large computational overhead that hinders its applicability. METHODS: We sought to develop an alternative method for CS reconstruction that only requires two gridding and one regridding operation in total, irrespective of the number of iterations. This proposed technique is evaluated on phantom images and whole-heart coronary MRI acquired using 3D radial trajectories, and compared to conventional CS reconstruction using CG algorithms in terms of quantitative vessel sharpness, vessel length, computation time, and convergence rate. RESULTS: Both CS reconstructions result in similar vessel length (P = 0.30) and vessel sharpness (P = 0.62). The per-iteration complexity of the proposed technique is approximately 3-fold lower than the conventional CS reconstruction (17.55 vs. 52.48 seconds in C++). Furthermore, for in-vivo datasets, the convergence rate of the proposed technique is faster (60±13 vs. 455±320 iterations) leading to a ∼23-fold reduction in reconstruction time. CONCLUSIONS: The proposed reconstruction provides images of similar quality to the conventional CS technique in terms of removing artifacts, but at a much lower computational complexity.

PURPOSE: To develop a novel pulse sequence for free-breathing, multislice, native myocardial T1 mapping. METHODS: The slice-interleaved T1 (STONE) sequence consists of multiple sets of single-shot images of different slices, acquired after a single nonselective inversion pulse. Each slice is only selectively excited once after each inversion pulse to allow sampling of the unperturbed longitudinal magnetization in the adjacent slices. For respiratory motion, a prospective slice-tracking respiratory navigator is used to decrease through-plane motion followed by a retrospective image registration to reduce in-plane motion. STONE T1 maps were calculated using both a two-parameter and three-parameter fit model. The accuracy and precision of the STONE sequence for different T1 , T2 , and inversion pulse efficiency were studied using numerical simulations and phantom experiments. T1 maps from 14 subjects were acquired with the STONE sequence and T1 s were compared to the MOdified Look-Locker Inversion recovery sequence (MOLLI). RESULTS: In numerical simulations and phantom experiments, the STONE sequence using a two-parameter fit model yields more accurate T1 times compared to MOLLI, with similar high precision. The three-parameter fit model further improves the accuracy, but with a reduced precision. The native myocardial T1 times were higher in the STONE sequence using two- or three-parameter fit compared to MOLLI. The standard deviation of the T1 times was lower in the STONE T1 maps with a two-parameter fit compared with MOLLI or a three-parameter fit. CONCLUSION: The STONE sequence allows accurate and precise quantification of native myocardial T1 times with the additional benefit of covering the entire ventricle. Magn Reson Med, 2014. © 2014 Wiley Periodicals, Inc.

PURPOSE: To compare accuracy, precision, and reproducibility of four commonly used myocardial T1 mapping sequences: modified Look-Locker inversion recovery (MOLLI), shortened MOLLI (ShMOLLI), saturation recovery single-shot acquisition (SASHA), and saturation pulse prepared heart rate independent inversion recovery (SAPPHIRE). MATERIALS AND METHODS: This HIPAA-compliant study was approved by the institutional review board. All subjects provided written informed consent. Accuracy, precision, and reproducibility of the four T1 mapping sequences were first compared in phantom experiments. In vivo analysis was performed in seven healthy subjects (mean age ± standard deviation, 38 years ± 19; four men, three women) who were imaged twice on two separate days. In vivo reproducibility of native T1 mapping and extracellular volume (ECV) were measured. Differences between the sequences were assessed by using Kruskal-Wallis and Wilcoxon rank sum tests (phantom data) and mixed-effect models (in vivo data). RESULTS: T1 mapping accuracy in phantoms was lower with ShMOLLI (62 msec) and MOLLI (44 msec) than with SASHA (13 msec; P < .05) and SAPPHIRE (12 msec; P < .05). MOLLI had similar precision to ShMOLLI (4.0 msec vs 5.6 msec; P = .07) but higher precision than SAPPHIRE (6.8 msec; P = .002) and SASHA (8.7 msec; P < .001). All sequences had similar reproducibility in phantoms (P = .1). The four sequences had similar in vivo reproducibility for native T1 mapping (∼25-50 msec; P > .05) and ECV quantification (∼0.01-0.02; P > .05). CONCLUSION: SASHA and SAPPHIRE yield higher accuracy, lower precision, and similar reproducibility compared with MOLLI and ShMOLLI for T1 measurement. Different sequences yield different ECV values; however, all sequences have similar reproducibility for ECV quantification.