High Field and BOLD Signal Behaviour

It has been clear from the very first experiences that the strength of the main magnetic field (B0) is a major factor in determining the amplitude of BOLD dependent fMRI signal changes [33] (Fig. 10.3). The physical reasons for the field dependence are inherent in the mechanisms producing the BOLD effect. In fact, the bulk magnetic susceptibility difference between blood containing paramagnetic deoxyhaemoglobin and surrounding diamagnetic tissue, on which difference the BOLD effect is based, increases with the main magnetic field strength, determining larger MR signal changes

Fmri Signals Motor

Fig. 10.3. Coronal fMRI maps of motor function obtained at 1.5 T (a) and at 3T (b) during a block design experiment with a button press at a 0.5 Hz frequency. The maps have been elaborated with Brain Voyager using a General Linear Model Analysis after undergoing a motion correction and a spatial smoothing pre-processing step. The map obtained at 3 T shows clear advantages in the spatial extension of the activation (sensitivity) particularly in the supplementary motor region, and a better correspondence of the activation with the anatomy of the motor system (specificity). The intensity of fMRI signal (BOLD %) doubles on passing from 1.5 and 3 T

Fig. 10.3. Coronal fMRI maps of motor function obtained at 1.5 T (a) and at 3T (b) during a block design experiment with a button press at a 0.5 Hz frequency. The maps have been elaborated with Brain Voyager using a General Linear Model Analysis after undergoing a motion correction and a spatial smoothing pre-processing step. The map obtained at 3 T shows clear advantages in the spatial extension of the activation (sensitivity) particularly in the supplementary motor region, and a better correspondence of the activation with the anatomy of the motor system (specificity). The intensity of fMRI signal (BOLD %) doubles on passing from 1.5 and 3 T

between baseline and activated states if high-field strength units are used [34]. As for the magnitude of the field dependence of the BOLD effect, it has been considered to vary from linear to quadratic [34-36], showing an inhomogeneous spatial distribution, principally depending on the relationship between dimension of voxels and dimension of blood vessels contained in them. Studies on the field dependence of the apparent transverse relaxation rate [36] showed that the BOLD effect is proportional to B0 if the voxels contain large vessels (venules and veins; d>10 |im), dropping to less than linear if vessels are larger than voxels. The relationship rises to more than linear in voxels containing a mixture of brain tissue, capillaries and ve-nules with a small diameter (less than voxel dimension), and reaches quadratic behaviour (B02) for voxels containing smaller vessels (capillaries and venules; d <10 |im). The biophysical model explaining this complex signal behaviour requires taking into account two basically different modalities of spin dephasing in the extravascular space, known as dynamic and static averaging regimes [35]. The dynamic averaging is caused by the diffusion of water molecules during the time span preceding echo readout, through the inhomoge-neous magnetic field determined by the presence of a blood vessel in a given voxel. The degree of inhomoge-neity is strongly conditioned by the vessel diameter. At a distance equal to the diameter of the vessel, the perturbation induced by the local magnetic field ( 0) is reduced to the 25% of the A<w0 measured at the vessel boundary. The maximum effect of the dynamic averaging regime will be experienced when vessel dimensions are in the range of the typical diffusion distances for a given echo time, so ranging between 2 and 10 |im in the common conditions posed by the fMRI experiment. Its effects will be visible in a spin-echo experiment through the changes in T2 [35].

For larger vessels (>10 |im), complete dynamic averaging for the entire voxel will not be possible.

The diffusion processes will generate only local dynamic averages over a subsection of the volume spanned by the magnetic field gradients produced by the blood vessel, in the so-called static averaging regime. A signal loss in the voxel will be in any case induced by the static averaging regime and will be visible in a gradient echo experiment in the absence of refocu-sing radiofrequency pulses [35].

The distinction between dynamic and static averaging regimes is fundamental to explaining the field dependence of BOLD with respect to vessel dimensions. In the regime of dynamic averaging, the BOLD effect is expected to vary as the square of the external magnetic field, while in the static averaging regime the dependence is linear. Even for the smallest vessels, the quadratic dependence of extravascular BOLD effect on the external magnetic field is not expected to persist forever with the increasing B0. At some very high external magnetic field strength, predicted by the theory but not

Fig. 10.4 a, b. High resolution fMRI of human auditory cortex obtained at high-field strength (7 T) (reprinted from [45] with permission). Despite the use of very small voxel dimensions, the 7 T EPI series presents a high enough signal to noise ratio to produce a very high „anatomical" detail of functional information

Fig. 10.4 a, b. High resolution fMRI of human auditory cortex obtained at high-field strength (7 T) (reprinted from [45] with permission). Despite the use of very small voxel dimensions, the 7 T EPI series presents a high enough signal to noise ratio to produce a very high „anatomical" detail of functional information yet verified experimentally, even the extravascular space surrounding the capillaries will behave as in the static averaging regime.

In any case, whatever the exact dependence on the field strength, the BOLD signal from a time-series at high magnetic field strength will gain inherent sensibility and specificity to microcirculation with the increasing B0 (Figs. 10.4,10.5). Interestingly, the possibility of operating at a higher spatial resolution will further enhance the spatial specificity of fMRI towards local neuronal activity either because of the greater weight of small vessel contribution and because of the greater capability to separate the large vessel contribution in an imaging voxel. The effects on fMRI spatial resolution can be remarkable. Paralleling the potential for reducing voxel dimensions due to the improved signal to noise ratio, the increasing contribution to the BOLD effect coming from the capillary bed can push the effective spatial resolution towards and beyond the ideal physiological limit of 1 mm2. In fact, if the initial dip is taken into account, its potential for improving spatial resolution converges with the improvement of signal to noise ratio and with the preferential sensitivity to capillary bed effects, further enhancing spatial resolution even below the millimetre range [8, 9].

Fig. 10.5. High resolution fMRI of human auditory cortex obtained at high-field strength (7 T) (reprinted from [45] with permission). Owing to the high spatial resolution it becomes possible to map brain functions at the level of tiny clusters within larger functional areas. In the auditory cortex, as an example, it is possible to single out the cortical location of the highest sensitivity to different frequencies of a tonal stimulation

Fig. 10.5. High resolution fMRI of human auditory cortex obtained at high-field strength (7 T) (reprinted from [45] with permission). Owing to the high spatial resolution it becomes possible to map brain functions at the level of tiny clusters within larger functional areas. In the auditory cortex, as an example, it is possible to single out the cortical location of the highest sensitivity to different frequencies of a tonal stimulation

Unraveling Alzheimers Disease

Unraveling Alzheimers Disease

I leave absolutely nothing out! Everything that I learned about Alzheimer’s I share with you. This is the most comprehensive report on Alzheimer’s you will ever read. No stone is left unturned in this comprehensive report.

Get My Free Ebook


Post a comment