An optimization method for designing SENSE imaging RF coil arrays

An optimization method in RF coil array design for SENSE imaging is described. Using this method the optimized RF coil geometries can be calculated numerically given the required SENSE imaging performance. Although this method can be applied to optimize the RF coil arrays for both 1D and 2D SENSE imaging, to demonstrate the potential applications of this method, we designed RF coil arrays for 2D SENSE imaging and compared their performance by simulation. An optimized 4-channel receive-only RF coil array designed for 2D SENSE imaging was implemented and tested to demonstrate the feasibility of the proposed technique. Imaging results showed reasonable agreement with the simulations, thus the method can be applied to RF coil array designs for SENSE imaging when optimum imaging performance is desired.     

Three-dimensional nonlinear vibration of gear pairs

This work investigates the three-dimensional nonlinear vibration of gear pairs where the nonlinearity is due to portions of gear teeth contact lines losing contact (partial contact loss). The gear contact model tracks partial contact loss using a discretized stiffness network. The nonlinear dynamic response is obtained using the discretized stiffness network, but it is interpreted and discussed with reference to a lumped-parameter gear mesh model named the equivalent stiffness representation. It consists of a translational stiffness acting at a changing center of stiffness location (two parameters) and a twist stiffness. These four parameters, calculated from the dynamic response, change as the gears vibrate, and tracking their behavior as a post-processing tool illuminates the nonlinear gear response. There is a gear mesh twist mode where the twist stiffness is active in addition to the well-known mesh deflection mode where the translational stiffness is active. The twist mode is excited by periodic back and forth axial movement of the center of stiffness in helical gears. The same effect can occur in wide facewidth spur gears if tooth lead modifications or other factors such as shaft and bearing deflections disrupt symmetry about the axial centers of the mating teeth. Resonances of both modes are shown to be nonlinear due to partial and total contact loss. Comparing the numerical results with gear vibration experiments from the literature verifies the model and confirms partial contact loss nonlinearity in experiments.     

Automatic tuned MRI RF coil for multinuclear imaging of small animals at 3T

We have developed an MRI RF coil whose tuning can be adjusted automatically between 120 and 128 MHz for sequential spectroscopic imaging of hydrogen and fluorine nuclei at field strength 3 T. Variable capacitance (varactor) diodes were placed on each rung of an eight-leg low-pass birdcage coil to change the tuning frequency of the coil. The diode junction capacitance can be controlled by the amount of applied reverse bias voltage. Impedance matching was also done automatically by another pair of varactor diodes to obtain the maximum SNR at each frequency. The same bias voltage was applied to the tuning varactors on all rungs to avoid perturbations in the coil. A network analyzer was used to monitor matching and tuning of the coil. A Pentium PC controlled the analyzer through the GPIB bus. A code written in LABVIEW was used to communicate with the network analyzer and adjust the bias voltages of the varactors via D/A converters. Serially programmed D/A converter devices were used to apply the bias voltages to the varactors. Isolation amplifiers were used together with RF choke inductors to provide isolation between the RF coil and the DC bias lines. We acquired proton and fluorine images sequentially from a multicompartment phantom using the designed coil. Good matching and tuning were obtained at both resonance frequencies. The tuning and matching of the coil were changed from one resonance frequency to the other within 60 s.     

Vibro-acoustic propagation of gear dynamics in a gear-bearing-housing system

This work developed a computational process to predict noise radiation from gearboxes. It developed a system-level vibro-acoustic model of an actual gearbox, including gears, bearings, shafts, and housing structure, and compared the results to experiments. The meshing action of gear teeth causes vibrations to propagate through shafts and bearings to the housing radiating noise. The vibration excitation from the gear mesh and the system response were predicted using finite element and lumped-parameter models. From these results, the radiated noise was calculated using a boundary element model of the housing. Experimental vibration and noise measurements from the gearbox confirmed the computational predictions. The developed tool was used to investigate the influence of standard rolling element and modified journal bearings on gearbox radiated noise.     

A double end-cap birdcage RF coil for small animal whole body imaging

Proper design of a birdcage coil plays a very important role in obtaining high-resolution small animal magnetic resonance imaging. The RF field homogeneity and the coil filling factor directly affect the signal-to-noise ratio (S/N) and therefore limit the resolution. It has been shown that a conductive end-cap placed on one side of the coil can improve the RF field inhomogeneity near this area. This also contributes to an increase in the S/N by reducing the length of the RF coil. While this is true near the end-cap, the distal half of the coil still suffers from poor homogeneity and S/N. Consequently, such a shortfall may hinder small animal whole body imaging. In order to improve the coil performance for a larger imaging volume, we designed a new small animal birdcage RF coil by adding a detachable second end-cap to the open end. The performance of single end and double end RF coils was compared experimentally. The results indicate that the double end-cap can provide superior uniformity along the long axis of the coil. Furthermore, if one wishes to obtain the same homogeneity within a given volume, a double end-cap would have less than half of the length of the single end-cap coil leading to a superior S/N performance.     

Fast imaging for magnetic resonance electrical impedance tomography

In magnetic resonance electrical impedance tomography (MREIT), currents are injected into an object, the resulting magnetic flux density is measured using MRI, and the conductivity distribution reconstructed using these MRI data. The relatively long acquisition times of conventional MREIT methods limit the signal averaging rate and are susceptible to motion artifacts. In this study, we reconstructed the conductivity distribution of an agarose gel phantom from data acquired in under a minute using a single-shot, spin echo, echo planar imaging (SS-SEPI) pulse sequence. The results demonstrate that SS-SEPI can be used for MREIT data acquisition.