Xinqiang Yan1,2, John C. Gore1,2,3, and William A. Grissom1,2,3
1Institute of Imaging Science, Vanderbilt University, Nashville, TN, United States, 2Radiology, Vanderbilt University, Nashville, TN, United States, 3Biomedical Engineering, Vanderbilt University, Nashville, TN, United States
Synopsis
At 7T and higher, the B1 fields of loop and microstrip coils become asymmetric. However, B1 fields of dipole antennas are still symmetric. The different behaviors of dipole and microstrip coils may
be explained by the fact that they have similar magnetic-field vectors but different Poynting vectors. We propose to
manipulate the Poynting vector and thus the symmetry of the B1 patterns of
microstrip coils using capacitor-segmented ground (CSG) planes. This concept has been validated
by numerical studies and practical MRI experiments. The CSG method provides
additional flexibility for manipulating the shape of the B1 field, which may be
advantageous for RF shimming and parallel transmission.Background
At field strengths of 7T and higher, the RF wavelength in
biological tissue becomes similar to human head and body dimensions. Under this
condition, the
B1 fields of loop and microstrip coils become asymmetric [1]. However,
the
B1 fields of dipole antennas are still nearly symmetric even in body
imaging at 7T [2]. The different behaviors of dipole and microstrip coils may
be explained by the fact that they have similar magnetic (
H-field) vectors but different Poynting vectors. We propose to
manipulate the Poynting vector and thus the symmetry of the
B1 patterns of
microstrip coils using capacitor-segmented ground (CSG) planes in order to
achieve improved coil performances.
Theory and Methods
In a conventional microstrip coil
[3, 4], the E-field
inside the coil
is perpendicular to the conductor and the ground plane, as shown
in Fig. 1A.
When the ground is segmented by capacitors (referred to as Cseg),
a parallel E-field component
is produced, as shown in Fig. 1B. This component rotates the Poynting vector toward the
target space and makes the B1 field pattern more symmetric, similar to a dipole antenna.
To validate this approach, we performed numerical
simulations using HFSS (Ansys, Canonsburg, PA, USA). A conventional microstrip coil was modeled
on a Teflon bar (25Χ5Χ2.5 cm3) (Fig. 1A, C). The widths of the conductor
and ground were 1 cm and 5 cm, respectively. The ground plane
of the CSG microstrip coil was uniformly
segmented by seven identical capacitors (Cseg) (Fig. 1B, D). Cseg was set with different values (150 pF, 68 pF, 33 pF, 15 pF and 7.5 pF)
to vary the length of the parallel E-field
component. A cylindrical phantom with 30 cm
length and 20 cm diameter was placed 2.5 cm below the coil. The phantom electrical properties were
б=0.6 S/m and ξr=55, approximating average tissue properties at 7T.
Input power was set at 1W in all
simulations.
We also built a conventional and CSG microstrip coils with Cseg=7.5 pF.
GRE images and B1+ maps on a water phantom were acquired on a 7T human Philips Achieva scanner (Philips Healthcare, Cleveland, Ohio, USA).
The parameters of the GRE sequence were: FA=30o,
TR/TE=100/10 ms, FOV=180 X 180 mm2, matrix= 256X256, thickness=3 mm.
B1+ maps were measured with the DREAM method [5].
Results
The Poynting vector distribution inside the
water phantom is shown in Fig. 2A (sagittal plane). The conventional coil’s Poynting vector points into the phantom only
in the region immediately adjacent to the coil.
In comparison, particularly for Cseg =15 and 7.5 pF, the Poynting
vector of the CSG coil was almost uniformly directed into the phantom. Figs. 2B
and C show the current distributions along the ground plane and the S11
plots vs. frequency. As Cseg decreases, its equivalent
impedance increases and thus current in the ground decreases, so the current
distribution becomes more like that of a dipole. Figs. 2D and E show the B1+ and
B1- fields, respectively. As expected, the B1 field pattern became
more symmetric (dipole-like) as Cseg decreases, thus demonstrating the
ability to manipulate the symmetry of the B1 patterns by tuning Cseg.
Fig. 3A
shows the constructed microstrip coils. For the CSG coil, three capacitors were
connected in parallel (2.4 pF+2.7pF+2.4pF=7.5pF) in each slit.
Fig. 3B compares the measured S11 curves and loaded Q-values. Figs.
3C and D show the GRE images and B1 maps, respectively. Similar to the
simulation results, the B1 patterns of two coils are quite different: the
conventional coil has a twisting B1 field and an obvious
lobe, while the CSG microstrip coil has a relatively symmetric B1 field.
Discussions and Conclusion
We have
developed a capacitor-segmented ground (CSG) approach for microstrip coils to
change their intrinsic
B1 field patterns. This concept has been validated by
numerical studies and practical MRI experiments. The CSG method provides
additional flexibility for manipulating the shape of the
B1 field, which may be
advantageous for RF shimming and parallel transmission. Although microstrip
arrays have relatively low coupling, strong cross-talk still exists in 16-channel
arrays [6, 7]. If the coupling among microstrip elements can be reduced using
different capacitor segmentation schemes between elements, it should provide an
easy and efficient method to create densely-spaced microstrip arrays. Furthermore,
the CSG microstrip coil can be seen as a transmission-line-folded dipole
antenna. However, unlike the dipole, the CSG microstrip coil can be fed at the
end as well as at the center and the selection of length is more flexible.
Acknowledgements
This work was supported by NIH R01 EB016695.References
[1] C.
M. Collins, et al, MRM. 65:1470–1482 (2011). [2] A.J. Raaijmakers, et al, MRM. 66: 1488-1497 (2011). [3]
G. Adriany, et al, MRM. 53:434–445 (2005). [4] X. Zhang, et al, MRM. 46:
443–450 (2001). [5] K. Nehrke, et al, MRM. 68:1517–26 (2012). [6] G. Adriany,
et al, MRM. 59: 590–597 (2008). [7] C. J. Snyder, et al, MRM. 67: 954–964
(2012).