MRI contrast using solid-state, B1-distorting, microelectromechanical systems (MEMS) microresonant devices (MRDs).

Presently, signal generation in MRI depends on the concentration and relaxivity of protons or other MR-active nuclei, and contrast depends on local differences in signal. In this proof-of-principle study, we explore the use of nonchemical, solid-state devices for generating detectable signal and/or contrast in vitro and in vivo. We introduce the concept of microresonant devices (MRDs), which are micron-sized resonators fabricated using microelectromechanical systems (MEMS) technology. Fifteen-micrometer (15-microm)-thick, coil MRDs were designed to resonate at the 3T Larmor frequency of protons (127.7 MHz) and were fabricated using tantalum (Ta) oxide thin-film capacitors and copper-plated spiral inductors. The performance of MRDs having final diameters of 300, 500, and 1000 microm were characterized in saline using a radio frequency (RF) scanning microscope and a clinical 3T MR scanner. The measured B(1) fields of 300 microm to 1000 microm MRDs ranged from 3.25 microT to 3.98 microT, and their quality factors (Q) ranged from 3.9 to 7.2. When implanted subcutaneously in the flank of a mouse, only MRDs tuned to the resonant frequency of protons generated a measurable in vivo B(1) field. This study lays the foundation for a new class of solid-state contrast agents for MRI.

Sodium MRI of the human kidney at 3 Tesla.

The sodium concentration gradient in the kidney (from the cortex to the medulla) serves to regulate fluid homeostasis and is tightly coupled to renal function. It was previously shown that renal function and pathophysiology can be characterized in rat kidneys by measuring the sodium gradient with (23)Na MRI. This study demonstrates for the first time the ability of (23)Na MRI to map the distribution of sodium in the human kidney and to quantify the corticomedullary sodium gradient. The study was performed on a 3T Signa LX scanner (GE) using an in-house-built quadrature surface coil. (23)Na images of volunteers were acquired using a 3D coronal gradient-echo sequence at a spatial resolution of 0.3 x 0.3 x 1.5 cm(3) in a 25-min scan time. The signal intensity (relative to the noise) increased linearly from the cortex to each of the medullae with a mean slope of 1.6 +/- 0.2 in relative arbitrary units per mm (Rel.u./mm, N = 6) and then decreased, as expected, toward the renal pelvis. Water deprivation (12 hr) induced a significant increase of 25% (P < 0.05) in this gradient. Based on these results, we suggest that sodium MRI can serve as a valuable noninvasive method for functional imaging of the human kidney.