Synopsis

A system has been developed to enable the acqusition of 16-channel ¹³C data on a Philips Achieva 7T scanner. Radiofrequency mixers are used to convert the transmitted signal to ¹³C, and the received signal back to ¹H to be acquired by the host system's receiver. A 16-element unilateral breast coil has been developed, and data have been acquired that show a substantial SNR improvement from use of the array.

Introduction

Studies showing correlations between the fatty acids in breast tissue and the risk of breast cancer make natural abundance ¹³C particularly attractive to use to assess triglyceride composition^1,2. Even at high fields, the low natural abundance and gyromagnetic ratio of ¹³C lead to inherent sensitivity limitations, making the use of array coils desirable^3,4.

Development and use of these coils is hindered by current capabilities of clinical scanners which commonly include up to 64 1H channels but only a single channel for X-nuclei. Frequency translation allows available ¹H channels to be used for any X-nucleus, providing a straightforward approach for circumventing this issue. This work describes the design and testing of a 16-channel ¹³C array coil and adaptable frequency-translation system for reception of 16 channels of ¹³C data through standard ¹H receive channels.

Methods

A 16-channel, unilateral, ¹³C breast receive array coil has been developed⁶. The design was based on Wald et al.’s “soccer-ball” geometry for decoupling and mounted on a 3D-printed, hemispherical ABS former with dimensions chosen to accommodate breast sizes of 80% of the American female population^5,6. Connection to a match/tune/decoupling network and modular preamplifier box⁷ provided an additional >14 dB of inter-coil decoupling while simultaneously matching each element to better than -20 dB. The preamplifier box contains low-pass filtering to avoid saturation by the ¹H transmit signals and routing lines for the -5V DC signal to activate a detuning trap which provided greater than 35 dB of active detuning to each receive element.

The receive array is surrounded by an actively-detunable ¹³C quadrature Helmholtz/saddle-pair transmit coil and a linear ¹H Helmholtz coil for shimming and scout imaging. Matching of all transmit coils to better than -20 dB was performed on-site during experimental setup

A 16-channel frequency translation system has been developed^8-11. A local oscillator (LO) is generated at the difference in frequency between ¹H and the X-nucleus. As shown in Figure 1, the ¹H transmit pulse is intercepted and mixed down to the frequency of the X-nucleus before the RF power amplifier. On the receive path, the X-nuclear signal from the coil is preamplified, mixed up to the ¹H frequency, and inserted into the host system’s ¹H receiver.

The system is divided into two main components, shown in Figure 2. The control unit (on the left), situated in the equipment room, controls and generates the LO, translates the RF transmit pulse, and supplies power to the 16-channel translation unit. The translation unit (on the right), situated adjacent to the magnet, uses non-magnetic active RF mixers to convert the preamplified received signal to the ¹H frequency. Figure 3 shows the ¹³C coil with a large spherical oil phantom used in the experiments.

Results

Experiments were conducted on a 7T Philips Achieva system. After ¹H shimming, the stock system was used to acquire spectra using the quadrature ¹³C volume coil for transmit and receive. This was used to calibrate the RF pulse before acquiring reference data.

After setting up the translation system, the transmit pulse was calibrated until the voltage and duration produced at the input of the transmit coil matched that produced by the reference scan. The frequency translation system was used to acquire data from the 16-element ¹³C array using the existing scanner 1H multi-channel receiver platform.

Discussion

Conclusion

The array coil and frequency-translation system demonstrated SNR increases for acquisition of natural abundance ¹³C spectra within the breast observed even when using only basic signal reconstruction techniques. Immediate future work includes localized spectroscopy using CSI, and testing the ability of coil elements to provide coarse localization of specific biomarker signatures through their reception profiles alone. A 3D-printed, compartmentalized phantom containing different NMR-active ¹³C chemicals (DMSO, TMS, Glycerol, and olive oil) over select receive elements has been built to test this.

This coil/translation system would also be appropriate for providing acceleration in metabolic DNP studies where SNR is sufficient but temporal resolution is more critical¹². Additionally, the flexibility of the frequency-translation setup allows it to be used for practically any other nuclei, as well as in cases where simultaneous acquisition of several nuclei is desirable.

Acknowledgements

Support from the Cancer Prevention and Research Institute of Texas through research grant RP150456 is gratefully acknowledged.

We greatly appreciate the assistance of Sandeep Ganji of Philips Healthcare in processing the raw spectroscopic data.

References

1. Bougnoux P, Giraudeau B, Couet C. Diet, cancer, and the lipidome. Cancer Epidem Biomarkers and Prev. 2006;15(3):416-421.

2. Chajes V, Thiebaut A, Rotival M, et al. Association between serum trans-monounsaturated fatty acids and breast cancer risk in the E3N-EPIC study. Amer Jour of Epidem. 2008;167(11):1312-1320.

3. Roemer P, Edelstein W, Hayes C. The NMR phased array. Mag Res in Med. 1990;16(2):192-225.

4. Wright S, Wald L. Theory and application of array coils in MR spectroscopy. NMR in Biomedicine. 1997;10(8)394-410.

5. Wiggins G, Triantafyllou C, Potthast A, et al. 32-channel 3 Tesla receive-only phased-array head coil with soccer-ball element geometry. Mag Res in Med. 2006;56(1):216-223.

6. Wilcox M, Ogier S, Sergey C, et al. A Sixteen-Channel Array Coil for Carbon-13 Spectroscopy of the Breast at 7T. ISMRM 2017:2656.

7. Reykowski A, Wright S, Porter J. Design of matching networks for low noise preamplifiers. Mag Res in Med. 1995;33(6):848-852.

8. Lee R, Gianquinto R, Constantindes C, et al. A broadband phased‐array system for direct phosphorus and sodium metabolic MRI on a clinical scanner. Mag Res in Med. 2000;43(2):269-277.

9. Ogier S, McDougall M, Wright S. Frequency Translation for 1H Decoupled Multichannel ¹³C Spectroscopy. ISMRM 2016:2171.

10. Ogier S, Wright S. Frequency Translation for 1H Decoupled Multichannel ¹³C Spectroscopy. ISMRM 2015:1786.

11. Ogier S, Wright S. A Frequency Translation Approach for Multichannel ¹³C Spectroscopy. IEEE-EMBC 2015:1564-1567.

12. Ardenkjær-Larsen J, Fridlund B, Gram A, et al. Increase in signal-to-noise ratio of> 10,000 times in liquid-state NMR. PN-AS. 2003;100(18): 10158-10163.