Jennifer Nussbaum1, Simon Gross1, David O. Brunner1, Christoph Barmet1,2, Thomas Schmid1, Benjamin E. Dietrich1, Markus Weiger1, and Klaas P. Pruessmann1
1Institute for Biomedical Engineering, University and ETH Zurich, Zurich, Switzerland, 2Skope Magnetic Resonance Technologies, Zurich, Switzerland
Synopsis
To measure
the spatiotemporal magnetic field evolution during MR procedures for image reconstruction
and real-time field control, best field probe performance is desired. We propose
improved field probes with 19F spherical droplet samples formed and positioned
with gelled deuterium oxide. It is shown
that these spherical samples have an isotropic k-space range and thus de-phase
less along the capillary than the common probes. Furthermore, with BIR-4 adiabatic plane rotation pulses the flip angle can
be perfectly adjusted, opening a new realm of field monitoring methods.Introduction
NMR field probes are
increasingly used to measure the spatiotemporal magnetic field evolution during
MR procedures
1,2. The sensor heads are typically formed by enclosing
a 1H or 19F NMR-active liquid in a capillary surrounded by a solenoid coil for
signal excitation and detection. The sensitive volume is shaped by the
solenoid's spatial transmit and receive characteristics. One key downside of
this approach is that the resulting effective droplet is only softly defined,
resulting in a suboptimal trade-off between signal yield and k-space range that
can be covered without probe de-phasing. What is more, the inhomogeneous transmit
field amplitude causes a flip angle gradient throughout the excitation volume,
which reduces the SNR. For optimal performance the sensitive volume should be
spherical, sharply delineated, and fully excited with uniform flip angle. In
the present work this is achieved by the transition to physical droplets
suspended in a gelled medium combined with adiabatic excitation. By using BIR-4
3,4
adiabatic plane rotation pulses the excitation flip angle can be exactly controlled.
Methods
19F NMR field probes with spherical sample
were built according to the schematic shown in Fig. 1. The sample is a
hexafluorobenzene droplet enclosed between D
2O plugs in a glass capillary
of either 1.3 mm or 0.8 mm inner diameter. To shape and immobilize the droplets
the plugs were gelled by adding 1% Imagel 69157 (Gelita AG, Germany). Their
magnetic susceptibility was matched to the droplets using MnCl
2∙4H
2O.
The sample liquids were doped with Cr(tmhd)
3 (48.7 mM) to reduce T
1.
To compare performance, conventional virtual droplet probes
2 were built using the same sample liquid and dimensions. Probe
excitation was performed alternatively with a 5 µs block pulse, a 48 µs
adiabatic half-passage (AHP, 90°) and a 48 µs BIR-4 (90°) adiabatic plane rotation pulse of
varying power. To include tolerance to small frequency offsets, hyperbolic secant (HSn)
adiabatic pulses were used
5. Measurements were performed at 3T in a Philips Achieva
MRI scanner and the 3D gradient echo images
of the droplet were acquired at 4.7 T using a Bruker
PharmaScan animal MRI scanner.
Results and Discussion
Figure
1 (right) illustrates the successful formation of highly spherical probe droplets as
confirmed by high-resolution MRI. Due to the spherical shape of the sample the probe signals exhibit the same
drop-off in all gradient directions (Fig. 2 left) which
is not the case for the virtual droplet probes (Fig. 2 right). With the physical droplet the signal decay under gradients along the capillary is less steep initially and steeper towards zero, resulting in substantially greater k-space range for any given signal threshold. The impact of
pulse choice is summarized in Fig. 3. We varied the B
1 amplitude of
the pulse and measured the initial absolute value of the FID. This results in a
curve proportional to the sine of the flip angle achieved at the corresponding
B
1 amplitude, meaning the peak value represents 90° flip angle.
Adiabatic excitation robustly yields maximum signal from sufficient B
1
amplitude upwards. Conventional block pulse excitation naturally suffers from
partial under- and over-flipping due to non-uniform B
1, making the
flip angle sensitive to the B
1 amplitude (Fig.3). Importantly,
the proposed physical droplet boundaries prevent expansion of the excited
volume and thus deterioration of k-space range with increasing adiabaticity
(Fig. 4). In the virtual droplet probe, on the other hand, enhancing B
1 enlarges the excited sample volume, which results in reduced k-space
range.
With
BIR-4 pulses any specified flip angle can be reliably induced as demonstrated in Fig. 5.
Conclusion
The trade-off between
signal yield and k-space range of NMR field probes can be optimized by forming
spherical fluorocarbon samples in a gelled polar medium. Sharp droplet
delineation permits the use of adiabatic pulses for uniform excitation
irrespective of gradient exposure. Perfect adjustment of the flip angle permits
experiments with smaller flip angles and faster re-excitation. Improved field
probe performance will be instrumental in the ongoing development of field
cameras for system diagnostics, image reconstruction, and real-time field
control.
Acknowledgements
No acknowledgement found.References
[1] De Zanche, N., Barmet, C., Nordmeyer-Massner, J.A., Pruessmann, K.P., 2008.NMR probes for measuring magnetic fields and field dynamics in MR systems. Magn. Reson. Med. 60, 176–186.
[2] Barmet, C., De Zanche, N., Wilm, B.J., Pruessmann, K.P., 2009. A transmit/receive system for magnetic field monitoring of in vivo MRI. Magn. Reson. Med. 62, 269–276.
[3] Staewen, R.S., Johnson,
A.J., Ross, B.D., Parrish, T., Merkle, H. and Garwood, M. 3D FLASH imaging
using a single surface coil and a new adiabatic pulse, BIR-4. Invest. Radiol.
25, 559-567 (1990).
[4] Garwood, M. and Ke, Y. Symmetric pulses to induce
arbitrary flip angles with compensation for RF inhomogeneity and resonance
offsets. J. Magn. Reson. 94, 511-525 (1991).
[5] Tannús, A., & Garwood, M. (1996). Improved Performance of
Frequency-Swept Pulses Using Offset-Independent Adiabaticity. Journal of
Magnetic Resonance, Series A, 120(1), 133–137.