T2*-weighted functional magnetic resonance imaging (fMRI) is used widely to investigate brain function, both at rest and in response to a stimulus or task. Unfortunately, BOLD fMRI is susceptible to spatial signal variations due to inhomogeneities in the static magnetic ?eld (B0) that are unrelated to neuronal activity. This effect is pronounced near air/tissue boundaries, where steep magnetic ?eld gradients can cause the BOLD signal to vanish. As a result, the BOLD signal is often partly or completely missing from the orbitofrontal cortex and ventromedial prefrontal cortex, believed to be important for high-level cognitive tasks such as decision-making and adaptive learning. To address this problem, we will develop a whole-brain 3D BOLD fMRI acquisition and reconstruction protocol with minimal signal loss artifacts and improved temporal SNR. We propose a three-pronged approach to reducing signal loss in 3D BOLD fMRI: We will (i) replace the conventional slab-selective radiofrequency (RF) excitation with a 3D spatially tailored RF pulse that ?pre-phases? the spins to make them refocus at a later time, (ii) perform model-based image reconstruction from parallel imaging data that accounts for the remaining effects of susceptibility-induced through-voxel B0 gradients, and (iii) perform z-shimming as needed to alleviate the pre-phasing demands placed on the RF pulse. We will also employ a more ef?cient radiofrequency (RF) spoiling scheme that enables reduced spoiler gradients. We hypothesize that the proposed protocol will achieve more complete spatial coverage than 2D fMRI, and more complete assessment of ventromedial prefrontal cortex (vmPFC) activation in obsessive-compulsive disorder (OCD) patients. Our goal is that the proposed protocol will serve as a broadly applicable ?drop-in replacement? for existing BOLD fMRI sequences, and that it will enable novel studies into the role of, e.g., the orbitofrontal cortex and ventromedial prefrontal cortex in resting-state and task-based fMRI.
Functional magnetic resonance imaging (fMRI) is used to measure brain function, and has revolutionized our under- standing of cognitive processes over the last 25 years. However, fMRI images suffer from signal ?dropout? artifacts near the air cavities in the skull, which makes it dif?cult to reliably assess brain function in, e.g., the orbitofrontal cortex. This is limiting our ability to investigate what cognitive role such ?hidden? brain regions may play. In this project we will develop an improved fMRI technique that we hope will enable true whole-brain fMRI.
