Simultaneous Multi-Slice Cardiac MR Multitasking for Motion-Resolved, Non-ECG, Free-Breathing T1–T2 Mapping

Abstract

The aim of this study is to simultaneously quantify T1/T2 across three slices of the left-ventricular myocardium without breath-holds or ECG monitoring, all within a 3 min scan. Radial simultaneous multi-slice (SMS) encoding, self-gating, and image reconstruction was incorporated into the cardiovascular magnetic resonance (CMR) Multitasking framework to simultaneously image three short-axis slices. A T2prep-IR FLASH sequence with two flip angles was designed and implemented to allow B1+-robust T1 and T2 mapping. The proposed Multitasking-SMS method was validated in a standardized phantom and 10 healthy volunteers, comparing T1 and T2 measurements and scan-rescan repeatability against corresponding reference methods in one layer of phantom vials and in 16 American Heart Association (AHA) myocardial segments. In phantom, Multitasking-SMS T1/T2 measurements showed substantial correlation (R2 > 0.996) and excellent agreement [intraclass correlation coefficients (ICC) ≥ 0.999)] with reference measurements. In healthy volunteers, Multitasking-SMS T1/T2 maps reported similar myocardial T1/T2 values (1,215 ± 91.0/41.5 ± 6.3 ms) to the reference myocardial T1/T2 values (1,239 ± 67.5/42.7 ± 4.1 ms), with P = 0.347 and P = 0.296, respectively. Bland–Altman analyses also demonstrated good in vivo repeatability in both the multitasking and references, with segment-wise coefficients of variation of 4.7% (multitasking T1), 8.9% (multitasking T2), 2.4% [modified look-locker inversion recovery (MOLLI)], and 4.6% (T2-prep FLASH), respectively. In summary, multitasking-SMS is feasible for free-breathing, non-ECG, myocardial T1/T2 quantification in 16 AHA segments over 3 short-axis slices in 3 min. The method shows the great potential for reducing exam time for quantitative CMR without ECG or breath-holds.

Free‐breathing, non‐ECG, simultaneous myocardial T1, T2, T2*, and fat‐fraction mapping with motion‐resolved cardiovascular MR multitasking

“…Nevertheless, it may still be time‐consuming if a short‐axis stack is desired in a clinical setting. Potential approaches for further acceleration include simultaneous multislice acquisition 66 and deep learning–based methods such as superresolution 67 . In this study, we only analyzed parametric maps and measurements in the diastolic phase.…”

Simultaneous multi-parametric acquisition and reconstruction techniques in cardiac magnetic resonance imaging: Basic concepts and status of clinical development

“…Like PI and CS, LRT methods work best with higher dimensional CMR applications (3D or more) since more redundancy exists at higher dimensions. For this reason, LRT methods have been successfully applied to create joint T1-T2 or T1-T2-cine images (19, 23,39,[79][80][81][82][83][84][85], exploiting the overlap between these contrasts. The adaptive and versatile nature of LRT methods have made them a major focus in the development of SMART (19,23,25,36,38,39,75,[79][80][81][82][83][84][85][86].…”

Artificial intelligence in cardiac magnetic resonance fingerprinting

“…Many simultaneous multiparametric approaches have recently been proposed to address these issues and provide co-registered multiparametric quantification, including Multitasking ( 6 , 7 ), steady-state techniques with multiparametric encoding ( 8 , 9 ), other free-running approaches ( 10 , 11 ) and Magnetic Resonance Fingerprinting (MRF), ( 12 ). MRF has the potential to provide not only multiple co-registered parametric maps in a time-efficient manner but can also include additional model corrections [e.g., B 0 ( 12 ), B 1 ( 13 ), slice profile ( 14 )].…”