Nuclear magnetic resonance (NMR) forms the basis of magnetic resonance imaging (MRI) and spectroscopy (MRS), both of which are frequently used non invasive methods to gain insight into human physiology and anatomy. NMR utilizes the magnetic properties of atomic nuclei of biological tissues and their interactions with a main static magnetic field (B0) and a time varying radio frequency (RF) field (B1).
For signal generation the RF field is irradiated by an RF coil, usually orthogonal to the main field direction. Spatial localization of the signal is then achieved via orthogonal field gradients. Magnetic resonance can be categorized as a low sensitivity method compared to other imaging methodologies like computer tomography (CT), or photon emission tomography (PET), due to the small number of signal producing nuclear spins. The available spatial and/or temporal resolution is often limited by the signal-to-noise ratio (SNR), which is proportional to the main magnetic field strength. As a rule of thumb: higher B0 facilitates higher SNR, at the cost of more complex field interactions. This causes a continued trend towards higher field strengths to enable finer spatial resolution or faster acquisition.
The resonance frequency is dependent on the type of nucleus investigated, e.g. 1H, 23Na, 31P, and increases linearly with the main field strength, resulting in shorter wave lengths and therefore more complex electromagnetic (EM) fields produced inside the human body by the RF coils in use. This complexity creates the need for specifically optimized designs for RF probes operating in the ultra high field regime (>3 T). For optimization purposes, knowledge of the EM field distribution and amplitude is needed. Since the MR signal is sensitive to the RF coil's magnetic B1 field, it can be visualized and quantified in the MR scanner. Unfortunately this is not true for the concomitant electric field (E). Although the E field is not needed for MR, it has to be carefully evaluated, since it deposits energy inside conducting samples which results in tissue heating, and therefore, being a major safety issue. Numerical simulation of Maxwell's equations enables the analysis of the produced EM fields before physically building a coil. In the last decades, EM simulations has proven to be an indispensable tool for RF coil design. This work describes a feasible workflow for the development of RF coils for transmission and reception of MR signals.
Special attention was paid to a comprehensive evaluation using numerical simulation, not only to optimize the design but also to ensure safety for future patient use. To assess the accuracy of the simulation results, various in-scanner validation methods are presented.
A 31P/1H RF coil conformed to the human calf for 31P metabolic investigations of skeletal muscle in the human lower extremities before, during, and after exercise, was designed, developed and tested according to the presented workflow. Studies conducted with the developed RF coil are presented shortly, to underline the functionality of the built system.