Introduction

Myocardial T1 and T2 mapping have been used to assess focal and diffuse myocardial pathologies. Current methods typically use a combination of ECG gating and breath-holding, although multi-dimensional methods such as MR Multitasking1 have shown promise for non-ECG and free-breathing T1-T2 mapping. However, Multitasking has so far been demonstrated for 2D single-slice¹ imaging in a 1.5 min scan and for 3D^2-3 volumetric imaging in 9 min or longer. Clinical protocols typically include T1 and T2 maps in several slices, but not in a full 3D stack, so neither the single-slice nor the 3D volumetric solutions are direct replacements for the conventional series of multiple T1 and T2 mapping scans. Here we combine the MR Multitasking framework with radial simultaneous multi-slice (SMS) acquisition^4-7 to perform cardiac-resolved T1/T2 mapping of basal, mid, and apical short-axis slices in a single non-ECG, free-breathing 3-min scan—roughly half the exam time of a conventional six-breath-hold sequence of three T1 slices followed by three T2 slices.

Methods

Sequence Design: The proposed continuous 2D sequence uses a radial trajectory modified to collect low-rank tensor auxiliary data interleaved with image data. A tiny golden angle (23.67°) angular increment was used between consecutive imaging readouts. A CAIPIRINHA-type phase modulation pattern was applied to the excitation pulses to achieve a multi-band factor of 3. The auxiliary data is acquired every other readout. Interleaved IR and T2-IR pulses are employed to generate the T1 and T2 contrasts (i.e., Fig. 1). The IR pulse is performed for the first recovery period, whereas the T2-IR pulses with four different preparation times (i.e., 30ms, 40ms, 50ms, 60ms) are performed in the following four recovery periods. The continuous acquisition sequence collects small-angle (i.e., 5°) FLASH readouts after the preparation pulses. Each FLASH excitation pulse was modified to have a different phase modulation so that it can excite 3 slices simultaneously with discrete Fourier slice encoding.
Image Reconstruction: The images acquired in the Multitasking framework can be represented as a 5-way tensor $$$ \boldsymbol{\mathcal{A}} $$$ with voxel location index $$$ \mathbf{r}=(x,y,z) $$$, a T1 recovery dimension, a T2 decay dimension, a cardiac motion dimension, a respiratory motion dimension. The high correlation along and across multiple time dimensions induces $$$ \boldsymbol{\mathcal{A}} $$$ to be low-rank, which can be factorized as $$$ \boldsymbol{\mathcal{A}}_{(1)}=\mathbf{U} \mathbf{\Phi} $$$, where $$$ \boldsymbol{\mathcal{A}}_{(1)} $$$ is the mode-1 matricization of $$$ \boldsymbol{\mathcal{A}} $$$, $$$ \mathbf{U} $$$ contains spatial coefficient maps, and $$$ \mathbf{\Phi} $$$ contains matricized, separable multi-dimensional temporal basis functions describing T1/T2 relaxation, cardiac motion, and respiratory motion¹. In reconstruction, $$$ \mathbf{\Phi} $$$ was first determined by the HOSVD of low-rank tensor–completed training data. $$$ \mathbf{U} $$$ was then recovered by fitting $$$ \mathbf{\Phi} $$$ to the acquired imaging data $$$ \mathbf{d} $$$, using the following optimization problem:$$ \mathbf{\hat{U}}=\text{arg}\underset{\mathbf{U}}{\operatorname{min}} ||\mathbf{d}-\Omega(\mathbf{FSU\Phi})||_2^2 + R(\mathbf{U}), $$ where $$$ \Omega $$$ is the undersampling operator, $$$ \mathbf{F} $$$ is the nonuniform Fourier transform operator, $$$ \mathbf{S} $$$ is the coil sensitivity operator, and $$$ R(\cdot) $$$ is an optional regularization functional to promote transform sparsity (chosen as a wavelet transform in this project).
T1,T2 Quantification: The native T1, T2 measurements are estimated from the signal model $$ s(A,B,T_1,T_2)=A \frac{1-e^{-TR/T_1}}{1-e^{-TR/T_1}cos(\alpha)} [1+(Be^{-\tau/T_2}-1)(e^{-TR/T_1}cos(\alpha))^n]\cdot sin(\alpha), $$ with amplitude factor $$$ A $$$, IR/T2-IR pulse efficiency $$$ B $$$, FLASH readout interval $$$ TR $$$, prescribed flip angle $$$\alpha$$$, and recovery time point $$$ n=1,2,..., N$$$, ( $$$ N $$$ is the number of FLASH pulses in a recovery period).

Results

T1 and T2 cine mapping images acquired from the 2D SMS Multitasking sequence are shown in Fig.2 and Fig. 3, along with the reference 2D diastolic and systolic T1, T2 images acquired from MOLLI and T2-prep FLASH sequences. Fig. 4 depicts the summary statistics for segment-wise T1/T2 values in 7 human volunteers. Multitasking T1 maps reported slightly lower myocardial T1 values (1117±39.7ms) than the MOLLI T1 measurements (1221±16.8ms). Multitasking T2 maps produced slightly higher myocardial T2 values (45.3±1.8 ms) than the T2-prep FLASH T2 values (38.6±1.4 ms). Fig. 5 shows the Bland-Altman plot of the multitasking segment-wise T1/T2 values and reference myocardial T1/T2 values.

Discussion and Conclusion

Co-registered T1 and T2 maps were simultaneously acquired at basal, mid, and apical slices using an SMS accelerated Multitasking technique in a 3-min scan. The myocardial T1, T2 quantification showed slight bias compared to reference MOLLI and T2-prep FLASH measurements, but were still in the reported normal range⁸. This method shows potential for reducing exam time for quantitative CMR without ECG or breath-holds. Future studies will focus on further reduction in scan time, validation in a larger cohort, repeatability assessments, and patient studies.

Acknowledgements

This work was supported by NIH 1R01EB028146.