Improved semi-LASER sequence with short echo time for ultra-high field using selective GOIA refocusing pulses
Michal Považan1,2, Lukas Hingerl1, Bernhard Strasser1, Gilbert Hangel1, Eva Heckova1, Stephan Gruber1, Siegfried Trattnig1,2, and Wolfgang Bogner1

1High Field MR Center, Department of Biomedical Imaging and Image-guided Therapy, Medical University Vienna, Vienna, Austria, 2Christian Doppler Laboratory for Clinical Molecular MR Imaging, Vienna, Austria

Synopsis

MR spectroscopy (MRS) profits from ultra-high field (UHF) with higher SNR and enhanced spectral resolution. However, the higher demand on bandwidth of RF pulses together with power limitations complicate the utilization of localization sequences such as PRESS or STEAM. A semi-LASER sequence appears to be a suitable candidate for UHF MRS if properly optimized. We aimed to implement selective GOIA refocusing pulses and optimize the gradient scheme to yield shortest echo time possible on a volume coil. Our semi-LASER sequence outperformed the conventional sequences in terms of SNR and chemical shift displacement artifact and proved to be applicable at UHF.

Introduction

Higher signal-to-noise ratio (SNR) and enhanced spectral resolution at ultra-high fields (UHF) allow shorter acquisition times or smaller detection volumes. However, the increased chemical shift dispersion sets a higher demand on the bandwidth (BW) of radiofrequency (RF) pulses to avoid excessive chemical shift displacement errors. Moreover, the power limitation of the RF pulses and gradients is more pronounced on UHF systems. Semi-localized by adiabatic selective refocusing (semi-LASER) sequences1,2 are a promising approach to overcome some of these UHF limitations. Semi-LASER sequences feature several advantages over other localization schemes. They possess lower chemical shift displacement errors (CSDE) than PRESS and do not sacrifice half of the signal as STEAM. In clinical practice, volume coils are used for brain MRI or MR spectroscopy (MRS). Previous semi-LASER implementations at 7T using volume coils were limited, due to the relatively low B1+ possible with volume coils, requiring either longer minimum echo times1 or utilization of B1+ field focusing3. We aimed to improve a semi-LASER sequence for use with volume coils at 7T with the shortest possible echo time (TE) to also detect strongly J-coupled metabolites.

Methods

For slice-selective excitation a Shinnar-Le-Roux-optimized 90° pulse was used. The conventional adiabatic full passage (AFP) refocusing pulses (duration 10ms, BW 2.5kHz)1 were replaced with shorter Gradient Offset Independent Adiabatic (GOIA) pulses (duration 3.5ms, BW 10kHz) based on WURST amplitude modulation with an order of W(16,4)4,5. The crusher gradients lengths and amplitudes were optimized for use at 7T. We have measured RF pulse profile of the GOIA-W(16,4) pulse. Phantom tests were carried out prior to in vivo measurements on a Magnetom 7T (Siemens Healthcare, Erlangen, Germany) scanner and 32-channel head coil (Nova Medical, Wilmington, USA) using a spherical phantom with a diameter of 16 cm that was filled with brain metabolites (NAA, creatine, choline, glutamate, lactate) in a phosphate-buffered solution at physiological pH and concentration. Conventional PRESS localization with Mao refocusing pulses (duration 6ms, BW 1.2kHz), STEAM with Hermite pulses (duration 4ms, BW 1.6kHz)6 and semi-LASER with GOIA refocusing were compared in terms of SNR. Three healthy volunteers were measured with VOIs (2x2x2cm3) placed in the parietal region and the following parameters: TR, 6 s; 32 averages; WET water suppression; EXOR phase cycling; readout BW 2.7 kHz; scan time 3:12 min; TEPRESS, 30ms; TESTEAM, 12ms; TMSTEAM, 8ms; TEs-LASER, 26ms.

Results

GOIA-W(16,4) provides excellent selection profiles even at short duration (Figure 1). The crusher gradient amplitudes were set to 15mT/m and 25mT/m (Figure 2) with a total length of 1ms to destroy any spurious signal. The minimal TE of 26ms was achieved with the given sequence parameters. The NAA amplitudes in vivo were 2.5 times higher for semi-LASER compared to STEAM and PRESS. In addition, the signal of strongly coupled glutamate was 2.3 times higher than glutamate measured with STEAM and 9 times higher than for PRESS (Figure 3). The SNR is summarized in Figure 4 for phantom and in vivo. The sequence parameters resulted in a CSDE of 50% for PRESS, 37% for STEAM and only 6% for semi-LASER over a 2 ppm range (i.e., Cr at 4ppm to NAA at 2ppm).

Discussion and Conclusion

GOIA pulses based on WURST modulation provide excellent profile features while maintaining high BW with shorter pulses and at lower RF power requirements. SNR is increased compared to other common localization techniques. The obstacles of MRS at UHF are observable in PRESS spectra (Figure 3), where the insufficient power of the pulses causes comparable signal intensities to STEAM and unbalanced gradients give rise to artifacts in 2-3ppm region (Figure 3, red arrows). Moreover, the low BW of pulses leads to huge CSDE. The semi-LASER sequence proved to be a well suitable localization technique for UHF capable of detecting strongly coupled systems due to inherent J-refocusing and reduced SNR loss due to T2 relaxation (i.e., signal decays with much slower T2rho during spin-locking). Overlapping of adjacent crusher gradients and implementation of asymmetric excitation pulses would further shorten the minimum TE of our semi-LASER sequence down to ~20ms.

Acknowledgements

No acknowledgement found.

References

1. Scheenen, T., Heerschap, A., Klomp, D., 2008. Towards 1 H-MRSI of the human brain at 7T with slice-selective adiabatic refocusing pulses. MAGMA 95–101. doi:10.1007/s10334-007-0094-y

2. Scheenen, T.W.J., Klomp, D.W.J., Wijnen, J.P., Heerschap, A., 2008. Short echo time 1H-MRSI of the human brain at 3T with minimal chemical shift displacement errors using adiabatic refocusing pulses. Magn. Reson. Med. 59, 1–6. doi:10.1002/mrm.21302

3. Boer, V.O., van Lier, a. L.H.M.W., Hoogduin, J.M., Wijnen, J.P., Luijten, P.R., Klomp, D.W.J., 2011. 7-T 1H MRS with adiabatic refocusing at short TE using radiofrequency focusing with a dual-channel volume transmit coil. NMR Biomed. 24, 1038–1046. doi:10.1002/nbm.1641

4. Andronesi, O.C., Ramadan, S., Ratai, E.-M., Jennings, D., Mountford, C.E., Sorensen, a G., 2010. Spectroscopic imaging with improved gradient modulated constant adiabaticity pulses on high-field clinical scanners. J. Magn. Reson. 203, 283–93. doi:10.1016/j.jmr.2010.01.010

5. Bogner, W., Chmelik, M., Andronesi, O.C., Sorensen, a G., Trattnig, S., Gruber, S., 2011. In vivo 31P spectroscopy by fully adiabatic extended image selected in vivo spectroscopy: a comparison between 3 T and 7 T. Magn. Reson. Med. 66, 923–30. doi:10.1002/mrm.22897

6. Gajdošík, M., Chadzynski, G.L., Hangel, G., Mlynárik, V., Chmelík, M., Valkovi, L., Trattnig, S., Kr, M., 2015. Ultrashort-TE stimulated echo acquisition mode ( STEAM ) improves the quantification of lipids and fatty acid chain unsaturation in the human liver at 7T 1283–1293. doi:10.1002/nbm.3382

Figures

RF profile of GOIA-W(16,4) pulse with 10kHz bandwidth and 3.5ms pulse length. GOIA pulses provide good selection profiles at 7T even for short pulse lengths.

Semi-LASER sequence with implemented GOIA refocusing pulses. Crusher gradients are placed around all four refocusing pulses, with varying amplitude (one pair 15mT/m, second pair 25mT/m). A further improvement is possible by overlapping adjacent crusher gradients and implementing asymmetric excitation pulse.

Brain phantom and in vivo comparison of Siemens PRESS (TE=30ms), STEAM (TE=12ms) and semi-LASER with GOIA pulses (TE=26ms). Red arrows depict the artifacts caused by unbalanced gradients in PRESS. Blue arrows show the strongly coupled signals of glutamate and myo-inositol.

SNR comparison for PRESS, STEAM and semi-LASER – Absolute SNR of NAA for phantom and in vivo measurement (calculated as a ratio of NAA amplitude to standard deviation of noise)



Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)
2388