Introduction

By the early 1980s MR imaging (MRI) was well on its way to establishing itself as a useful tool for diagnosis of a number of disorders, especially of the central nervous system (CNS) (Atlas, 2002). The reason for the rapid and spectacular success of MRI is the superb soft tissue contrast that imaging water distribution and relaxation times afford. In addition, the non-invasive nature of the technology makes it possible to readily test efficacy. By the end of the 1980s, the great success in generating anatomical images of normal and diseased tissue led a number of groups to seek ways to get functional information from MRI. Indeed, by the early 1990s techniques had been developed that enabled various aspects of tissue function to be assessed. Most important has been blood oxygen level dependent (BOLD) contrast, which enables detection of changes in hemoglobin oxygenation during regional activation of the brain (Kwong et al., 1992; Ogawa et al., 1992). BOLD-based MRI has rapidly grown into a technique that readily enables brain mapping during complex cognitive tasks (Moonen and Bandettini, 1999). Another class of MRI techniques sensitizes images to changes in diffusion of water in tissue (Wesbey et al., 1984) as well as quantifying preferred diffusion directions (Basser et al., 1994). Diffusion-weighted MRI has grown into an important tool for monitoring tissue damage due to ischemia in brain (Warach, 2002), increasing MRI sensitivity to white matter (WM) disorders (Ahrens et al., 1998) and for mapping fiber orientation in the brain (Mori et al., 2000). Another important way to add functional information to MRI is the class of techniques that enable regional measurement of tissue blood flow or perfusion.