Dynamic mapping of the human visual cortex by high-speed magnetic resonance imaging.

Abstract

We report the use of high-speed magetic resonance imaging to follow the changes in image intens in the human visual cortex during stimulation by a flashing checkerboard stimulus. Measurements were made in a 2.1-T, 1-m-diameter magnet, part of a Bruker Biospec spectrometer that we had programmed to do echo-planar imaging. A 15-cm-diameter surface coil was used to transmit and receive signals. Images were acquired during periods of stimulation from 2 s to 180 s. Images were acquired in 65.5 ms in a 10-mm slice with in-plane voxel size of 6 x 3 mm. Repetition time (TR) was generally 2 s, although for the long flashing periods, TR = 8 s was used. Voxels were located onto an inversion recovery image taken with 2 x 2 mm in-plane resolution. Image intensity increased after onset ofthe stimulus. The mean change in signal relative to the prestimulation level (AS/S) was 9.7% (SD = 2.8%, n = 20) with an echo time of 70 ms. Irrespective of the period of stimulation, the increase in magnetic resonance signal intensity was delayed relative to the stimulus. The mean delay measured from the start of stimulation for each protocol was as follows: 2-s stimulation, delay = 3.5 s (SD = 0.5 s, n = 10) (the delay exceeds stimulus duration); 20-to 24-s stimulation, delay = 5 s (SD = 2 s, n = 20).Functional mapping of the brain by nuclear magnetic resonance (NMR) methods has recently been demonstrated by two techniques. The first method uses a bolus of paramagnetic contrast agent (gadolinium/diethylenetriaminepentaacetic acid) as a tracer for blood volume (1). The agent is injected into the arm and produces variations in local blood volume magnetic susceptibility as the bolus passes through the capillary bed of the brain. The changes in susceptibility are monitored by using a transverse relaxation time (T2)-weighted, fast magnetic resonance-imaging sequence, such as echo-planar imaging (EPI) (2) or fast low-angle shot (FLASH) (3). This method is subject to the same assumptions as all kinetic tracer methods (4), but within these restrictions it gives a quantitative result. With this method Belliveau et al. (5) have made measurements in the human visual cortex and demonstrated an increase in blood volume during stimulation. The main drawback of the technique is the need for at least two injections of contrast agent to measure the resting and activated brain states.The second method relies solely on the paramagnetic effects of hemoglobin. In the oxygenated state (HbO2), the molecule is diamagnetic, but once dissociated from oxygen (deoxyhemoglobin; Hb), it becomes paramagnetic. Thus, changes in Hb concentration will also result in magnetic susceptibility changes in the capillaries. Complications may arise because the paramagnetic agent is not evenly distributed throughout the blood but is compartmentalized inside the erythrocytes. Ogawa et al. (6) have studied the effect of inhaled 02 content on the intensity of gradient-echo images ofrat brain at 7 T. They have shown that a decrease in oxygen content produces dark streaks in the i...

Influence of EEG electrodes on the BOLD fMRI signal

“…Most of the existing fMRI data has been collected using gradient echo (GE) sequences [Bandettini et al, 1992;Belliveau et al, 1991;Blamire et al, 1992;Frahm et al, 1992;Kwong et al, 1992;Ogawa et al, 1992;Turner et al, 1993]. This technique, however, is very Output of the Phase-Only Filter for the retinotopic maps with/without EEG electrodes (Fig.…”

In vivo17O NMR approaches for brain study at high field

“…[18][19][20] These physiological parameters can change during brain activation, and they provide the main signal sources detected by two of the most popular neuroimaging modalities: positron emission tomography (PET) [24][25][26][27] and functional magnetic resonance imaging (fMRI). [28][29][30][31][32][33][34][35] In general, cerebral bioenergetics underlying brain function is not readily accessible noninvasively, and how we probe it (technically) can influence our findings (the outcomes). 36 For instance, many conclusions resulting from in vitro or ex vivo studies may not be valid in the living brain.…”

Disparity of activation onset in sensory cortex from simultaneous auditory and visual stimulation: Differences between perfusion and blood oxygenation level‐dependent functional magnetic resonance imaging

“…Recently, both BOLD-and perfusion-based techniques were shown to be capable of detecting transient cerebral hemodynamic response associated with short stimulus durations (12)(13)(14)(15). With careful data processing, event-related (ER) BOLD response latencies as small as 100 msec after stimulus onset have been detected, despite relatively long repetition times (16).…”