Date of Award


Degree Type


Degree Name

Doctor of Philosophy


Medical Biophysics


Diffusion-weighted magnetic resonance imaging utilizes the magnetic gradients of the system to de-phase protons undergoing diffusion with respect to the overall mag­ netization. Areas of the image with reduced signal when compared to an un-weighted image represent where protons have undergone diffusion. The stronger the gradient applied during diffusion-weighting, the larger the signal loss due to diffusion, and the larger the b-value differentiating the diffusion coefficients. However, the maximum gradient strength during image acquisition is limited by both the original strength of the signal and peripheral nerve stimulation.

Nerve stimulation is induced because the changing magnetic fields of the gradient pulse sequence induce electric fields that cause stimulation. The stimulation thresh­ old can be measured either in terms of the pulse sequence parameters of maximum gradient strength and slew rate, or in terms of the induced electric field and the duration of the electric field pulse.

A finite-difference simulation was used to approximate the electric field induced inside a visible man model. The effect of varying the size, resolution, and position of the model inside the simulation was investigated with the wire pattern from a customized head/neck gradient coil. For accurate simulations, it was most important to ensure that the resolution of the model was sufficient to capture the air cavities of the sinus and trachea.


The peripheral nerve stimulation thresholds of a planar gradient coil were deter­ mined from human experiments. While the electrical stimulation threshold parame­ ters did not vary significantly from previous studies, the minimum gradient change and slew rate required to cause stimulation were significantly higher for the planar gradient than for reported thresholds of cylindrically designed gradient systems.

Several non-cylindrical localized gradient designs were investigated for diffusion- weighted contrast as a fourth gradient, in addition to the three imaging axes. Both resistive and inductive merits were investigated. Of these, inductive values proved to be the limiting factor when designing coils sized to perform in a full body MRI system. Optimal merit and gradient strength were obtained from a butterfly design, and planar coils provided localized strength over a larger region.

A butterfly coil was constructed with hollow copper wiring and powered to produce diffusion weighting during MRI. Diffusion contrast b—1300 s/mm2was obtained using the insert with significant time and signal to noise ratio improvements



To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.