Transceive Stripline Arrays for Ultra High Field Parallel Imaging Applications

G. Adriany, P. Van de Moortele, F. Wiesinger, P. Andersen, J. Strupp, X. Zhang, C. J. Snyder, W. Chen, K. P. Pruessmann, P. Boesiger, J. T. Vaughan, K. Ugurbil Center for MR Research, University of Minnesota Medical School, Minneapolis, United States, Institute for Biomedical Engineering, University and ETH Zurich, Switzerland We present four and eight channel transceive stripline arrays for ultra high field parallel imaging applications. Good coil decoupling between stripline array elements was achieved without preamplifier decoupling. With the four channel array and the eight channel array we achieved high reduction factors and excellent average gfactors. Our results confirm the prediction that the maximal achievable reduction factor increases with field strength. Introduction At ultra high fields, many applications that require EPI or SPIRAL type fast imaging, such as fMRI, are expected to significantly benefit from parallel imaging techniques due to reduction in time for acquisition of k-space data. Furthermore, it is also expected that the maximal feasible reduction factor will increase with higher field strength [1,2]. However, there are a number of challenges that arise for designing dedicated parallel imaging coils at ultra high fields. Strong coupling to the sample at the high frequencies mediates interactions between the separate coils, making it difficult to decouple them. There are image inhomogeneities caused partly by interaction with conductive tissue and partly by the differences between reception and excitation profiles [3]; these inhomogeneities aid parallel imaging strategies but pose problems for constructing coil arrays that cover the brain. Furthermore, the geometrical constraints for a experimental setup that includes separate transmit coil and receive array coils, as well as task presentation for fMRI studies, are severe. Transceive arrays seem particular advantageous in addressing these issues [4] and they additionally allow for transmit SENSE [5,6]. Here we present four and eight channel transceive array coil designs for parallel imaging of the brain at 7T and results obtained with these coil arrays. Methods We built four and eight channel transceive surface coil arrays according to stripline transmission line principles [7-9]. The four channel transceive coils were 13 cm x 12 cm in size with 2.5 cm inter coil spacing and oriented in 0,90,180 and 270 degree position around an elliptical former. The eight channel coils were built from 7cm wide and 14 cm long elements and evenly spaced with 1.5 cm inter coil spacing (Fig.1). All coils were built using 1cm wide copper tape for the coil conductors and 2cm wide copper tape for the ground conductor. All grounds were cut in one position to avoid eddy currents. Imaging experiments were performed on a 7 Tesla Magnet (Magnex Scientific, UK) equipped with a Varian console (Palo Alto, CA) and Siemens gradient amplifier (Erlangen, Germany). We utilized a single 4 kW RF amplifier (CPC, Brentwood, NY) and split the RF power 4-ways or 8-ways. The transmit phase increments for each channel were adjusted for optimal image homogeneity. T/R switches with low insertion loss of 0.2dB in each transmit path blocked transmitter noise during reception and enabled the used of low noise preamplifiers. Results and Discussion Utilizing stripline surface coils we were able to achieve good coil decoupling without preamplifier decoupling. No resonance peak split was observed and the coils could be tuned and matched for each subject. By adjusting the transmit phase and amplitude for each array coil independently we were able to achieve excellent parallel imaging performance while obtaining additionally RF shimming capabilities. Since the profile of the coils is less than 1cm there is sufficient space for task presentation hardware. The average Q0/ QL ratio of the coils was measured to be 285/70, indicating good transmit efficiency. Decoupling in the four channel coil between neighboring coils when loaded with a human head was -18dB and -35dB between opposing coils. We achieved similar values in the eight channel coil when using decoupling capacitors as an additional decoupling method. We compared the transceive arrays with our standard TEM head coils and acquired B1 field maps. A 90deg flip angle in the center of the head with 1kW RF power required a 350μs long square pulse with the four channel array, which is comparable to a volume coil [10]. Peripheral areas however are more efficiently excited with the transceive array and we were able to achieve spin inversion in areas of the lower brain where this previously was not feasible with a quadrature volume coil. Using full k-space acquisition, sum-of-squares from the four channels displayed good homogeneity in image intensity. Using SENSE, we were able to achieve acceleration factors of 3 with the four channel array with an average geometry factor of 1.41 when reducing phase encoding along the long axis and 1.65 when reducing along the short axis (Fig.3). The eight channel array achieved an excellent g-factor of 1.28 for rate 4 reduction indicating that the achievable reduction factor increases indeed with field strength.