Author's School

Graduate School of Arts & Sciences

Author's Department/Program



English (en)

Date of Award

January 2010

Degree Type


Degree Name

Doctor of Philosophy (PhD)

Chair and Committee

Mark Conradi


This thesis is divided into two parts. The first part examines diffusion in a number of human lung samples, while the second deals with diffusion in the more simplified environment of a single capillary tube. In the lung study we examine various types of lung samples: healthy and diseased) at a variety of diffusion lengths. Diffusion Magnetic Resonance Imaging: MRI) has been a very useful tool in the study of lungs and lung disease, particularly diffusion MRI using hyperpolarized gases, such as helium or xenon. However, since the polarized gas is difficult to prepare and, once prepared, suffers decay of the nonrenewable polarization: limiting both the number and strength of rf pulses that may be applied to the sample), most studies only measure diffusion at a handful of diffusion times, at most. Thus no extensive study has been performed to discover at what diffusion time: and hence diffusion length scale) the healthy and diseased lung differ most. By using fixed samples from lungs and gas at Boltzmann polarization, this study seeks to remedy that gap in knowledge. The immobile samples and continually renewed thermal polarization of the gas allowed signal averaging to make up for some of the signal lost by not using polarized gas, although the small signal-to-noise ratio did not allow for diffusion images. Instead, spectroscopic diffusion measurements were made, effectively treating the whole lung sample: approximately a 2 inch long by 1.34 inch diameter cylinder) as a single voxel. Because this study seeks to explore many different diffusion lengths, two gases were used, one with a large free diffusion coefficient, used to explore long length scales, and one with a smaller free diffusion coefficient, used to explore shorter length scales. In order to combine the measurements from both types of gas, the restricted diffusion: D/D0: measured diffusivity divided by the free diffusion of the gas used) is reported. T1 in each sample was used to calculate the free gas diffusion coefficient; the bulk gas measurements used to establish this T1 - diffusion relationship are reported as well. In the capillary study, we examine the signal decay due to diffusion in a single cylinder, for short diffusion times: lightly restricted diffusion). The signals are well-modeled by a sum of two exponentials, despite the single compartment housing the spins. The results agree with a previous theoretical examination of the problem. The implication for biological systems is that multiple decay signal components may not correspond to multiple physical compartments, despite the fact that multi-exponential decays in diffusion experiments are often taken as evidence for spins in multiple distinct compartments.


Permanent URL: