Improved data acquisition and processing strategies for blood oxygenatlon level-dependent (BOlO)-conlrast funcllona1 magnetie resonaneo imaging (fMRI), wllich enhance the functional eontrast-to-nolse ratio (CNR) by sampllng multiple echo limes In a single shot, are descrlbed. The dependence of Ihe CNR on Ti, Ihe image encoding time, and Ihe number of samplod oeho titnes are Investigated for exponentia1 fitting, echo summation, welghled echo summation, and averaglng of corrolalion maps oblainod at different echo limes. The mothod is validated In vlvo using visual stimulation and turbo proton echoplanar speelroseopie imaging (turbo-PEPSI), a new single-shot multi-slice MR spoclroscoplc Imaging teehnlque, whlch acqulres up 10 12 consocutive ochoplanar images wlth echo limes ranging from 1210213 msec. Quantitative Ti-mapplng slgnificanUy increasos Ihe measured extent of aetivatJon and the mean correlalion coefficient compared wilh convenlional echoplanar imaging. The sensltlvity gain with echo summation, wllicll is compulationally efficiet:'lt provides similar sensitivity as fitting. For all data processing methods sensltivlty is optimum wh on echo limes IIp 10 3.2 T 2 are sampled. This molhodology has implications for comparing functional sonsitivity at different magnetie field strengths and between braln regions with different magnetic field inhomogeneitics.

The sensitivity of functional magnetic resonance imaging (fMRI) in visual cortex to graded hypo-and hypercapnia was quantified in 10 normal subjects using single-shot multiecho echoplanar imaging (Turbo-PEPSI) with eight equidistant echo times (TEs) between 12 and 140 ms. Visual stimulation was combined with controlled hyperventilation and carbon dioxide inhalation to perform fMRI at six levels of end-expiratory pCO 2 Functional MRI (fMRI) is widely used to map brain activation with a spatial-temporal resolution that is unparalleled by other brain imaging techniques. However, fMRI suffers from a lack of quantification, which is in part due to complex signal dependence on a number of vascular factors, including blood flow, blood volume, blood oxygenation, and vascular architecture. This can complicate the between-subject and within-subject comparisons. Recently, the relationship between changes in blood oxygenation-level dependent (BOLD) contrast signal and cerebral blood flow (CBF) has been investigated by a number of research groups, using graded stimuli. In the visual cortex several groups reported a linear relationship between signal changes in echo-planar imaging (EPI) and regional CBF (rCBF) (1). Sadato et al. (2) compared fMRI and positron emission tomography (PET) using a graded finger-tapping paradigm. While fMRI signal changes in the primary sensorimotor cortex increased linearly with tapping rate, rCBF changes measured by PET tended to saturate at higher tapping rates. In a recent study of passive listening, a nonlinear relationship between the word presentation rate and the EPI signal change, with a tendency to saturate towards increasing presentation rate, was found in the auditory cortex (3). By contrast, relative CBF changes measured by PET were linearly dependent on presentation rate.BOLD contrast is also dependent on global CBF (gCBF), which may change during anxiety-and drugrelated changes in respiration. We and others have shown that decreases in gCBF during hyperventilation decrease baseline BOLD signal and fMRI contrast during visual stimulation (4). In the visual cortex we found an almost complete loss of functional contrast at end-expiratory pCO 2 (PETCO 2 ) ϭ 20 mm Hg (4). During moderate hypercapnia induced by CO 2 breathing, several groups reported increases in baseline signal and in activationinduced signal changes in visual cortex (5,6). By contrast, in previous PET studies, relative rCBF changes during visual stimulation were found to be independent of gCBF (7).Since EPI-based fMRI measures only relative changes in signal intensity at a fixed echo time (TE), changes in baseline T* 2 and in initial signal intensity (S o ) due to inflow effects may influence fMRI sensitivity and confound comparison of signal changes at different PETCO 2 levels. Calibration of EPI signal changes between subjects remains difficult. Absolute quantification of T* 2 and S o is thus desirable. Recently, we and others developed single-shot multiecho EPI methods which enable fitting of the water relaxation tim...

No abstract