Synopsis

Myocardial T1 mapping provides quantitative tissue characterization for the assessment of various cardiomyopathies. However, currently available myocardial T1 mapping techniques still have several limitations such as insufficient coverage, low spatial resolution, and the need of acquiring the data under multiple breath-holds. To overcome these problems, here we propose a free-running (free-breathing, no ECG gating) 3D whole heart T1 mapping technique with high isotropic spatial resolution. This approach allows for myocardial T1 mapping at arbitrary cardiac phases, enabling high-resolution dynamic T1 maps. The feasibility of the proposed sequence was validated against conventional methods in phantom and five healthy subjects.

Introduction

Methods

Sequence Design: The proposed IR-prepared 3D radial sequence is shown in Fig.1A. The sequence continuously acquires radial spokes conforming to the 3D golden angle distribution (6) to achieve pseudo-uniform distribution of spokes binned into arbitrary cardiac phases and T1 contrasts. The sequence was implemented on a 1.5T Philips MR scanner. Relevant scan parameters were: FOV=200mm³, spatial resolution=1.5mm³, TR/TE/flip angle=11.6ms/5.1ms/6°, water-excitation, scan time=9.5min.

Motion Binning and Reconstruction: Superior-inferior 1D respiratory motion was estimated from the k-space center of all spokes (Fig. 1B, C), and used to bin the k-space data into 5 equally populated respiratory phases (Fig. 1D). This self-navigated respiratory signal was validated against the corresponding respiratory bellow signal (Fig. 2A). The ECG signal was recorded and used to bin the acquired data into different cardiac phases. 3D translational respiratory motion was estimated from the intermediate reconstruction of low-resolution respiratory bin images at diastole cardiac phase (Fig. 2B). Respiratory motion correction was performed by correcting the phase of the k-space data using the estimated 3D translation motion parameters. The respiratory motion corrected k-space data was then binned into different T1 contrasts and cardiac phases. To reconstruct T1 image series at specific cardiac phase, parallel imaging, dictionary-based low-rank inversion (4), which exploits temporal compression along the contrast dimension, and patch-based reconstruction (5), which exploits local and non-local redundancy of 3D patches, were combined. The optimization problem can be formulated as an unconstrained Lagrangian formulation: $$L\left(I,\alpha,Y\right)=argmin\parallel EI-K \parallel_2^2+\lambda\parallel \alpha \parallel_0+\mu\parallel I-P\alpha-Y \parallel_2^2$$where $$$I$$$ denotes the compressed image series; $$$E=AU_{r}FS$$$ is the encoding operator, with $$$S$$$ being sensitivity maps, $$$F$$$ being Fourier Transform, $$$U_{r}$$$ being the low rank operator obtained by truncating the singular value decomposition of the dictionary by Bloch simulation, $$$A$$$ being the k-space sampling operator; $$$K$$$ is the undersampled data; $$$P$$$ is the patch grouping operator and $$$\alpha$$$ are the associated sparse coefficients; $$$Y$$$ is the Augmented Lagrangian multiplier; $$$\lambda$$$ controls the sparsity regularization contribution and $$$\mu$$$ is the penalty parameter. The above equation can be efficiently solved by operator-splitting via alternating direction method of multipliers (ADMM) (5). A dot product matching between the reconstructed singular value images and the dictionary was performed to generate the T1 maps.

MR Imaging: Phantom experiments were performed to test the accuracy of the proposed sequence using the 2D IR spin echo (IR-SE) sequence as gold standard. For the in vivo study, five healthy subjects (2 females, 29.2±3.3 years) were scanned with the proposed free-running 3D T1 mapping and a conventional 2D breath-hold, ECG-gated MOLLI sequence (7). Imaging was performed in the short axis orientation, with MOLLI imaging slice positioned at mid-ventricle, whereas the proposed sequence covered the whole heart.

Results

Discussion

Acknowledgements

References

1. Perea RJ, Ortiz-Perez JT, Sole M, Cibeira MT, de Caralt TM, Prat-Gonzalez S, Bosch X, Berruezo A, Sanchez M, Blade J. T1 mapping: characterisation of myocardial interstitial space. Insights Imaging 2015;6(2):189-202.

2. Kellman P, Hansen MS. T1-mapping in the heart: accuracy and precision. J Cardiovasc Magn Reson 2014;16:2.

3. Messroghli DR, Moon JC, Ferreira VM, Grosse-Wortmann L, He T, Kellman P, Mascherbauer J, Nezafat R, Salerno M, Schelbert EB, Taylor AJ, Thompson R, Ugander M, van Heeswijk RB, Friedrich MG. Clinical recommendations for cardiovascular magnetic resonance mapping of T1, T2, T2* and extracellular volume: A consensus statement by the Society for Cardiovascular Magnetic Resonance (SCMR) endorsed by the European Association for Cardiovascular Imaging (EACVI). J Cardiovasc Magn Reson 2017;19(1):75.

4. Asslander J, Cloos MA, Knoll F, Sodickson DK, Hennig J, Lattanzi R. Low rank alternating direction method of multipliers reconstruction for MR fingerprinting. Magn Reson Med 2018;79(1):83-96.

5. Bustin A, Ginami G, Cruz G, Correia T, Ismail TF, Rashid I, Neji R, Botnar RM, Prieto C. Five-minute whole-heart coronary MRA with sub-millimeter isotropic resolution, 100% respiratory scan efficiency, and 3D-PROST reconstruction. Magn Reson Med 2018.

6. Chan RW, Ramsay EA, Cunningham CH, Plewes DB. Temporal Stability of Adaptive 3D Radial MRI Using Multidimensional Golden Means. Magnetic Resonance in Medicine 2009;61(2):354-363.

7. Ugander M, Oki AJ, Hsu LY, Kellman P, Greiser A, Chen MY, Bandettini WP, Aletras AH, Arai AE. Myocardial Extracellular Volume Imaging by MRI Quantitatively Characterizes Myocardial Infarction and Subclinical Myocardial Fibrosis. Circulation 2010;122(21).