Céline Marquet1,2,3, Jihye Jang1,2,4, Andrew J. Powell1,2, and Mehdi H. Moghari1,2
1Department of Pediatrics, Harvard Medical School, Boston, MA, United States, 2Department of Cardiology, Boston Children's Hospital, Boston, MA, United States, 3Department of Informatics, Technical University Munich, Munich, Germany, 4Philips Healthcare, Boston, MA, United States
Synopsis
We developed a novel MRI pulse sequence RAS-Q
T1T2 for the simultaneous quantification of T1
and T2 of the myocardium using transient bSSFP imaging with a variable
flip angle scheme. RAS-Q T1T2 was systematically analyzed
based on a numerical simulation, as well as phantom and patient studies. We
show that RAS-Q T1T2 yields accurate T2
estimates for the myocardium with a trade-off in precision compared to
state-of-the-art methods, reducing scan time to less than 4s. The estimated T1
values reveal lower accuracy and precision than clinically established methods
and need further improvements.
Introduction
Longitudinal and transverse relaxation times (T1 and T2)
have long been used for tissue characterization. More recently, there has been
a surge of interest in measuring the T1 and T2 of the
myocardium. Increased non-contrast myocardial T1 values have been
related to various cardiomyopathies.1 Additionally,
increased myocardial T2 values have been shown to correlate with
myocarditis2 and edema3. The conventional
approach to assessing myocardial T1 and T2 is to apply
separate acquisition methods for each parameter. Several methods to
simultaneously quantify myocardial T1 and T2 have been
proposed.4,5,6,7 These techniques,
however, entail certain limitations in clinical applicability, such as long breath-hold
acquisition periods and scan times, which can be demanding for very young or
ill cardiac patients. To avoid such restrictions, we developed a novel pulse
sequence to simultaneously quantify the T1 and T2 of the
myocardium in less than 4s.Methods
The proposed RAS-Q T1T2
pulse sequence combines transient balanced steady-state free precession (bSSFP)
imaging,8,9,10 and a variable flip
angle scheme.11,12 Four images are acquired
with variable flip angles (FAs), (10°, 50°, 90°, 130°) at a specific cardiac
phase in 4 heartbeats (Figure 1). Assuming heartrates between 60-120 bpm, the
acquisition time is 2-4s. To account for partially recovered magnetization from
previous images, we introduce an additional, iterative term $$$\delta_i$$$
to the transient
phase bSSFP relaxation model10:
$$ M_{RASQ}\ (\alpha_i\ )=\delta_i\times M_{trans}\ (\alpha_i\ ), \\\text{where } M_{trans}(\alpha_i) =(M_0\sin(\frac{\alpha_i}{2})-M_{SS}(\alpha_i))\lambda(\alpha_i)^n+M_{SS}(\alpha_i) \\
\text{with } M_{SS}(\alpha_i) = M_0 \frac{\sqrt{E_2}(1-E_1)\sin(\alpha_i)}{1-(E_1-E_2)\cos(\alpha_i)-E_1E_2} \text{, }\\
\lambda(\alpha_i)=E_1\cos^2(\frac{\alpha_i}{2})+E_2\sin^2(\frac{\alpha_i}{2}) \text{, }E_{1/2}=exp(\frac{-T_R}{T_{1/2}}),\\
\text{and } \delta_1=M_0 \text{, } \delta_i=M_0-(M_0-M_{RASQ} (\alpha_{i-1})\cot(\frac{\alpha_{i-1}}{2}))exp(\frac{-t}{T_{1}}) $$
FAs are denoted as
$$$\alpha_i$$$ for
each respective image, $$$t$$$ describes
the time between heartbeats, $$$T_R$$$
is the repetition time, and $$$n$$$ was
fitted to match the underlying magnetization. To reduce the number of unknown
parameters, we normalize the pixel intensities by dividing them by the corresponding
pixel of the last image. This eliminates the term for proton density $$$M_0$$$
.
We systematically
investigated the feasibility of the novel RAS-Q T1T2 approach with a numerical
simulation, as well as a phantom and in
vivo study. The phantom study was performed to compare the proposed
sequence to T1 and T2 of the gold-standard spin echo (SE)
(T1 Acquisition: TR/TE:10000/11ms, 17 inversion
times: 50-5000ms (min-max), FA: 90°; T2 Acquisition: TR/TE:
10000/32*10ms,
FA: 90°), and Modified Look-Locker IR (MOLLI)13 (TR/TE:
2.53/1.27ms, FA: 35°, scheme: 5(3s)3, total acquisition time (TAT): 11s) for T1
evaluation, and T2-Prep14 (TR/TE:
2.65/1.33ms, FA: 35°, rest period: 2 heartbeats, T2-Prep delay: 0,
25, 50, and 75ms, TAT: ≤12s) for T2 evaluation. The experiments were
performed on the T1MES phantom15 at 4 different
heartrates (60, 80, 100, 120 bpm) with FOV: 200 (RL) × 200 (AP) × 10 (FH) mm,
CS-SENSE 3, and 10 startup pulses for all methods. We performed an in vivo feasibility study in 2 patients
for RAS-Q T1T2.
All scans were performed on a 1.5T Philips scanner (1.5T Achieva d-stream). The quantification algorithms were
implemented in MATLAB (MathWorks Inc., Natick, MA, USA). Method evaluation was
based on accuracy (mean error compared to SE) and precision (pixel-wise
standard deviation within region-of-interest). A paired student t-test was used for
statistical analysis and a p-value≤0.05 was assumed statistically significant. Results
For the phantom study, the
gold-standard SE yielded the following ranges for T1 = [257-1547ms] and
T2 = [48-235ms]. T1 quantification with MOLLI
demonstrated an accuracy/ precision of -35.8ms/ 9.8ms compared to SE. The
comparison of RAS-Q T1T2
to SE for T1 resulted in an accuracy/ precision of -155.5ms/
120.6ms. T2 quantification with T2-Prep demonstrated an
accuracy/precision of 12.3ms/ 3.9ms compared to SE. The comparison of RAS-Q T1T2 to SE
for T2 resulted in an accuracy/ precision of -9.4ms/ 12.5ms (Table
1). Figure 2 shows the RAS-Q T1T2 magnetization curves
for the phantom study at heartrate 60 bpm. Figure 3 shows the T1 and
T2 maps of the phantom for all sequences. Figure 4 demonstrates the T2
maps of 2 patients calculated with RAS-Q
T1T2.Discussion
RAS-Q T1T2 allows rapid myocardial T1
and T2 quantification in 4 heartbeats. RAS-Q T1T2 shows lower accuracy and
precision compared to MOLLI. RAS-Q T1T2 shows higher accuracy and lower precision
compared to T2-Prep-based T2 mapping, which is considered
a trade-off for a much shorter scan time and simultaneous estimation of T1.
Depending on the patient’s heartrate, the acquisition time of RAS-Q T1T2 is
2.75-6 times faster than for MOLLI or T2-Prep.Conclusion
We demonstrate technical feasibility of myocardial T1 and T2
mapping with RAS-Q T1T2.
Clinical applicability of T1 quantification needs to be investigated
further. The estimation of myocardial T2 in the phantom results in
accurate values with acceptable precision compared to T2-Prep. Future
studies to improve T1 quantification to establish RAS-Q T1T2 as a
method to simultaneously quantify T1 and T2 are warranted.Acknowledgements
This project was supported by a scholarship of the Harvard Scholarship Foundation Germany e.V.References
1.
Moon JC, Messroghli D, Kellman P, et
al. Myocardial T1 mapping and extracellular volume quantification: a Society
for Cardiovascular Magnetic Resonance (SCMR) and CMR Working Group of the
European Society of Cardiology consensus statement. Journal of Cardiovascular
Magnetic Resonance. 2013;15(1):92-92.
2.
Spieker M, Haberkorn S, Gastl M, et
al. Abnormal T2 mapping cardiovascular magnetic resonance correlates with
adverse clinical outcome in patients with suspected acute myocarditis. Journal
of Cardiovascular Magnetic Resonance. December 2017;19.
3.
Nishii T, Kono AK, Shigeru M, et al.
Cardiovascular magnetic resonance T2 mapping can detect myocardial edema in
idiopathic dilated cardiomyopathy. The International Journal of Cardiovascular
Imaging. April 2014;30.
4.
Santini F, Kawel-Boehm N, Geiser A,
Bremerich J, Bieri O. Simultaneous T1 and T2 quantification of the myocardium
using cardiac balanced-SSFP inversion recovery with interleaved sampling
acquisition (CABIRIA). Magnetic Resonance in Medicine. 2015;74:365-371.
5.
Akcakaya M, Weingärtner S, Basha TA,
Roujol S, Bellm S, Nezafat R. Joint myocardial T1 and T2 mapping using a
combination of saturation recovery and T2-preparation. Magnetic Resonance in
Medicine. September 2016;76(3):888-896.
6.
Qi H, Bustin A, Cruz G, et al.
Free-running simultaneous myocardial T1/T2 mapping and cine imaging with 3D whole-heart
coverage and isotropic spatial resolution. Magnetic Resonance Imaging.
2019;63:159-169.
7.
Hamilton JI, Jiang Y, Chen Y, et al.
MR fingerprinting for rapid quantification of myocardial T1, T2, and proton
spin density. Magnetic Resonance in Medicine. 2017;77(4):1446-1458.
8.
Oppelt A, Graumann R, Barfuss H,
Fischer H, Hartl W, Schajor W. FISP: eine neue schnelle Pulssequenz fuer die
Kernspintomographie. Electromedica. 1986;54(1):15-18.
9.
Bieri O, Scheffler K. Fundamentals
of balanced steady state free precession. Journal of Magnetic Resonance
Imaging. 2013;38:2-11.
10.
Scheffler K. On the Transient Phase
of Balanced SSFP Sequences. Magnetic Resonance in Medicine. April
2003;49(4):781-783.
11.
Worters P, Hargreaves B. Balanced
SSFP transient imaging using variable flip angles for a predefined signal
profile. Magnetic Resonance in Medicine. November 2010;64(5):1404-1412.
12.
Srinivasan S, Ennis DB. Variable
flip angle balanced steady-state free precession for lower SAR or higher
contrast cardiac cine imaging. Magnetic Resonance in Medicine. March
2014;71(3):1035-1043.
13.
Messroghli D, Radjenovic A, Kozerke
S, Higgins DM, Sovananthan MU, Ridgway JP. Modified Look-Locker inversion
recovery (MOLLI) for high-resolution T1 mapping of the heart. Magnetic
Resonance in Medicine. June 2004;52(1):141-146.
14.
Giri S, Chung YC, Merchant A, et al.
T2 quantification for improved detection of myocardial edema. Journal of
Cardiovascular Magnetic Resonance. December 2009;11:56.
15.
Captur G, Gatehouse PD, Kellman P,
et al. A T1 and ECV phantom for global T1 mapping quality assurance: The T1
mapping and ECV standardisation in CMR (T1MES) program. Journal of
Cardiovascular Magnetic Resonance. 2016;18(1):1-3.