Functional and metabolic MRI techniques are rapidly evolving and have tremendous potential for clinical brain disorder research. A significant effort has been made to determine the inter-subject and intra-subject reliability and reproducibility of fMRI rapid imaging techniques and protons spectroscopic imaging. The blood oxygenation level dependent (BOLD) contrast method, dynamic contrast methods, and arterial spin tagging approach have been used to obtain information about brain function. The BOLD method depends upon detection of changes in the signal intensity in response to an activation paradigm and is related to the changes in local deoxyhemoglobin concentration, blood flow and volume, local hematocrit and others within a region of the parenchyma. Dynamic contrast enhanced studies utilize rapid scan techniques and a bolus injection of a parmagnetic contrast agent which alters the local magnetic fields within the tissue and provides information on the relative cerebral blood volume (rCBV), mean transit time (MTT), arrival times (AT) and relative cerebral blood flow (rCBF). The arterial spin tagging approach involves the noninvasive labeling using radiofrequency pulses of water protons in the arteries and arterioles away from the acquisition slice to quantitate cerebral blood flow (CBF). Validation studies using BOLD techniques have demonstrated significant variation in the absolute number of activated pixels for a specific functional anatomical region of the brain when one compares the intra-subject and inter-subject results for experiments performed on the same or different days. The cause of this variability requires further evaluation, although, respiratory and cardiac motion, as well as, macroscopic head motion all contribute to problems with fMRI. Normalization for the total number of activated pixels for the subject, results in highly significant specificity, sensitivity and within subject reliability of whole-brain fMRI methods for brain mapping in control subjects. Currently, whole brain fMRI studies are performed with cognitive activation during working memory tasks with emphasis on possibly being able to distinguish control populations from patients with serious mental illnesses.
