Synopsis
In
hyperpolarized gas MRI, the accurate flip angle calibration and T1
measurements are important. Traditional flip angle calibration methods are
time-consuming and suffer from polarization losses during T1
relaxation. In this study, we propose a method to simultaneously calibrate flip
angles and measure T1 in vivo
during a breath-hold time of less than 4 seconds. The so-called single-breath method
is magnitude based and simple to use. We demonstrate the accuracy of this
method and contrast it with traditional methods. The results of the calibration
verified that it is robust and repeatable.Purpose
In
hyperpolarized gas MRI, the accurate calibration of flip angles is important for
maximizing the SNR per unit time[1].
However, the calibration of flip angles is sensitive to relaxation effects. Moreover,
the spin-lattice relaxation time (T
1) of the gas can reflect the
partial oxygen pressure, which is an important parameter related to the
efficiency of gas exchange in the lung[2]. The measurement of T
1
also requires a pulse flip angle calibration[3]. Most previous methods generally require more than two breaths: one for the flip angle
calibration and another one for the T
1 measurement. A previous
study[4] by Max
et al. provided a method for the simultaneous estimation
of T
1 and the flip angle in a single scan through acquisition at
non-regular time intervals. However, the long acquisition time needed for
mounting time intervals limits its usefulness during
in
vivo applications. In this work, we propose a novel method to
simultaneously calibrate flip angles and measure T
1 of
hyperpolarized gas during a short single breath-hold
in vivo.
Methods
Our proposed method for calibration in a single-breath is shown in Fig. 1a. First, we set
a transmitter gain TG
1 to get a small flip angle θ
1 in the range 3°-6°. After 8 excitations by θ
1, the transmitter gain is
changed to a smaller value TG
2 to obtain a larger flip angle θ
2.
Subsequently,
we set TG
3~TG
m to obtain gradually larger flip angles θ
3~θ
m, 8 excitations for each
angle. After acquiring all of the transverse signals by the N
excitations, we can fit the unknown flip angles using one of two methods. The
first method involves fitting the m angles to $$$S_{k}=S_{1}\cdot\left(\cos\theta\cdot\exp\left(-TR/T_{1}\right)\right)^{k-1}$$$ with k=1-8,
by setting an initial T
1 value, where TR is repetition time. Once
the value θ
1 is fitted, the flip angle θ
2 can be fitted to $$$\theta_{2}=\arctan\left[\tan\theta_{1}\cdot\exp\left(TR/T_{1}\right)\cdot S_{9}/S_{8}\right]$$$. We denote the flip angle θ
2 calibrated
from the first method as θ
2′, and the one calibrated using the second method as θ
2″.
If θ
2′ is equal to θ
2″, this means that the initial T
1
value is correct. If θ
2′ is not equal to θ
2″, this means
that the initial T
1 value is incorrect, and we should change the
T
1 to another value. The correct T
1 value can be derived from the m angles to minimize the measurement error. Then the
correct flip angles can be fitted with the correct T
1. All
experiments were performed on a 7 T animal MRI scanner with a
homebuilt 8-leg rigid transmit-receive birdcage coil. The traditional multi-breath
method was used for contrast. In the traditional flip angle measurement
experiment, 16 different flip angles are measured with 16 breaths, and 112
excitations for each angle. T
1 is measured by varying the delay
time between trigger and the 1st RF excitation, with 7 different
delay times in 7 breaths. The normal breaths are ventilated with pure oxygen, where different T
1 values are acquired by using different xenon gas pre-wash times.
Both
the flip angle and T
1 are measured in a single breath-hold by the method described in this study. The single-breath
method was repeated 5 times for testing its robustness. TR was 67 ms for balloon and 33.7 ms for rat. The breath-hold time
needed was 4 seconds for rat.
Results
The
result of flip angle calibration is shown in Table 1. The
parameters were those in $$$TG_{\theta}=TG_{90°}+A\cdot\log_{10}{90/\theta}$$$. Xenon pre-wash time was
0 in all measurements. T
1 was set to 14 s in balloon, and 5 s in
rat, when calibrated by multi-breath method. The result of T
1 measurement
is shown in Table 2, where the traditional method was repeated twice for the
0-time pre-wash in balloon, and once for others.
Discussion
The
calibration results of both flip angles and T
1 by the single-breath
method described in this study are accurate. As shown in Table 1, the TG
90° in rat lung is smaller than that in balloon, which
is because of the RF stacking in rat tissues. The parameter A is close to 20, but not exactly equal
to 20, which is an interesting phenomenon. The measured T
1 values
with 0-time pre-wash are shorter than those with the 1-time pre-wash, which
reflects different oxygen pressures. The deviations of 5 measurements are small,
thus confirming the robustness of this method.
Conclusion
This
study proposed a novel method to simultaneously calibrate flip angles and
measure T
1 of hyperpolarized gas during a short single breath-hold
in vivo.
Acknowledgements
No acknowledgement found.References
[1] Zhao L et al. JMR
Series B 1996;113(2):179-183. [2] Rizi RR et al. MRM 2004;52(1):65-72. [3] Moller HE et al. MRM 2001;45(3):421-430. [4]
Puckeridge M et al. JMR
2012;222:68-73.