Assessing the Feasibility of Dynamic 31P Spectroscopy for Metabolic Studies With a 1.0T Extremity Scanner.

OBJECTIVE: The feasibility of conducting in vivo non-localized 31P Magnetic Resonance Spectroscopy (MRS) with a 1.0T extremity scanner and the potential to increase accessibility of this important diagnostic tool for low cost applications is revisited. METHODS: This work presents a custom transmit-only quadrature birdcage, four-element receive coil array, and spectrometer interfaced to a commercial ONI 1.0T magnet for enabling multi-channel, non-1H frequency capabilities. A custom, magnetic resonance compatible plantar flexion-extension exercise device was also developed to enable exercise protocols. The coils were assessed with bench measurements and 31P phantom studies before an in vivo demonstration. RESULTS: In pulse and acquire spectroscopy of a phantom, the array was found to improve the signal-to-noise ratio (SNR) by a factor of 1.31 and reduce the linewidth by 13.9% when compared to a large loop coil of the same overall size. In vivo testing results show that two averages and a four second repetition time for a temporal resolution of eight seconds was sufficient to obtain phosphocreatine recovery values and baseline pH levels aligned with expected literature values. CONCLUSION: Initial in vivo human skeletal muscle 31P MRS allowed successful monitoring of metabolic changes during an 18-minute exercise protocol. SIGNIFICANCE: Adding an array coil and multinuclear capability to a commercial low-cost 1.0T extremity scanner enabled the observation of characteristic 31P metabolic information, such as the phosphocreatinerecovery rate and underlying baseline pH.

Feasibility of Using a 1T Extremity Scanner with a Four-Element Array to Detect 31P in the Human Calf.

The feasibility of conducting in vivo non-localized skeletal muscle 31P Magnetic Resonance Spectroscopy (MRS) with a low-cost extremity 1 Tesla magnet is demonstrated. We designed and built a transmit-only quadrature birdcage, four-element receive coil array, and employed a home-built spectrometer interfaced with a commercial ONI 1.0T magnet. In phantom comparison tests with a large loop coil of comparable size, the array was found to improve the SNR by a factor of 1.8 and the linewidth from 0.72 ppm to 0.45 ppm. Phantom and in vivo testing results show only 6 averages with a 4 second repetition time are required to obtain quantifiable 31P spectra. Initial in vivo human skeletal muscle 31P spectra successfully allowed for peak characterization. A low-cost approach to MRS could enable more widespread use of this tool in clinical diagnosis and in vivo metabolic research.

Flexible RF Filtering Front-End For Simultaneous Multinuclear MR Spectroscopy.

Simultaneously interrogation of multiple nuclei has been of interest since the very earliest days of MRI [1]-[3]. Our group and several others are revisiting this topic [4]-[6]. Very fast broadband electronics make it possible to digitize a wide spectrum, including multiple nuclei, but this places great demands on data throughput. Another issue is that there can be great variance between RF preamplifier gain required for the different nuclei. To overcome the data problem, it is desirable to use undersampling, but this requires passband filtering around the resonant frequency of each nuclei. Here we present a frequency agile front end that provides separate data paths for each nucleus, either from a single coil or from multiple ports, allows independent gain, filters each using very flexible transmission line filtering, and then combines them back for undersampling.

An Adjustable-Length Dipole Using Forced-Current Excitation for 7T MR.

Ultrahigh field imaging of the body and the spine is challenging due to the large field-of-view (FOV) required. It is especially difficult for RF transmission due to its requirement on both the length and the depth of the ${\rm{B}}_{1}^{{\rm + }}$ field. One solution is to use a long dipole to provide continuous current distribution. The drawback is the natural falloff of the ${\rm{B}}_{1}$ field toward the ends of the dipole, therefore the ${\rm{B}}_{1}^{{\rm + }}$ per unit square root of maximum specific absorption rate ${\rm{(B}}_{1}^{{\rm + }}{\rm{/ \surd SAR}}_{{\rm{max}}})$ performance is particularly poor toward the end of the dipole. In this study, a segmented element design using forced-current excitation and a switching circuit is presented. The design provides long FOV when desired and allows flexible FOV switching and power distribution without additional power amplifiers. Different element types and arrangements were explored and a segmented dipole design was chosen as the best design. The segmented dipole was implemented and tested on the bench and with a phantom on a 7T whole body scanner. The switchable mode dipole enabled a large FOV in the long mode and improved ${\rm{B}}_{1}^{{\rm + }}{\rm{/ \surd SAR}}_{{\rm{max}}}$ efficiency in a smaller FOV in the short mode.