| Liver fibrosis is a chronic condition that involves the build up of an excessive amount of scar tissue in the liver parenchyma and is associated with excess deposits of collagen, proteoglycans, and other macromolecules in the extracellular matrix. In addition, liver fibrosis causes alterations in hepatic blood flow. Liver fibrosis culminates in liver cirrhosis, which is incurable. The current gold standard for the assessment of liver fibrosis is biopsy, which is invasive and can lead to potentially serious complications. Therefore, noninvasive methods to accurately diagnose fibrosis are needed. Diffusion weighted imaging (DWI), which detects the random Brownian motion of water molecules and is capable of determining biologic characteristics of tissue through the application of large imaging gradients, has the potential to fulfill this need.;The current standard for DWI in cancer imaging is to collect data at two diffusion weightings (b-values) to calculate an apparent diffusion coefficient (ADC). The standard calculation of ADC involves the underlying assumption that the DWI signal decreases monoexponentially as a function of b-value. However, the DWI signal is more complex and is affected by perfusion. The IVIM theory allows for the extraction of perfusion information from this complex diffusion signal through the collection of many b-values.;Liver fibrosis is also associated with a stiffening of the liver. Magnetic Resonance Elastography (MRE) measures the propagation of acoustic shear waves, generated by a passive driver, through liver tissue and gives estimates of liver stiffness.;This dissertation focuses on applying the advanced DWI methods developed in the brain to the liver. Specifically, DWI methods were optimized and used to noninvasively diagnose liver fibrosis. In addition, the ability of MRE to diagnose liver cirrhosis was evaluated. First, IVIM-DWI was optimized for use in the liver. Simulations were run to determine the b-value distribution that minimized the error of IVIM parameters. Low b-values (01000s/mm2) had little to no effect.;DWI in the liver poses problems not seen with DWI in the brain. The liver is sensitive to artifacts from breathing and and/or cardiac induced motion. In an attempt to compensate for these effects, liver images were collected within a single breath hold (BH), using respiratory triggering (RT), or during free breathing (FB). A set of eight control subjects was imaged with the goal of determining the best set of triggering techniques and diffusion directions for DWI in the liver. The RT scans with three orthogonal diffusion directions acquired simultaneously (3in1) proved the most repeatable. Further analysis also revealed that very low b-values (0 |
