Biomedical Engineering Reference
In-Depth Information
follows is a brief overview of the development of fMRI and some of the re-
search into what it actually samples of brain activity.
There are now a number of different fMRI techniques that can be used for
making movies of focal changes in brain physiology related to one or a set of
targeted mental functions. These techniques can be roughly categorized by the
use of exogenous (injected) contrast agent versus methodologies based upon
intrinsic contrast (natural contrast agent in the blood, or the effects of flow on
MRI signal). Functional MRI with contrast agents was first demonstrated by
Belliveau et al. (20), using echo-planar imaging in combination with the para-
magnetic contrast agent gadolinium, bound to a chelating agent, DTPA. This
general methodology works because the presence of Gd-DTPA within the par-
enchymal vasculature increases the decay rate of the MR signal (1/
T
2
) in a re-
gionally specific fashion. This, in turn, changes the image contrast, and serial
measurement of image intensity can be converted to regional cerebral blood
volume. If injections of Gd-DTPA are made during different experimental con-
ditions—for instance, the rest condition of no movement and the targeted condi-
tion of finger apposition—contrasting images acquired during each experimental
condition can lead to a measure of cerebral blood volume change associated
with the experimental perturbation to the system. The cerebral blood volume
change can be evaluated statistically, and overlaid on a structural MRI to illus-
trate the anatomical localization. When using novel contrast agents with long
blood half-lives, as can be employed in animal models, this technique becomes
particularly powerful. Repeated injections are no longer required, and changes in
blood volume can be assessed dynamically throughout the entire experiment.
Moreover, the dose of injected agent can be dialed to the optimum value in order
to produce the strongest of all possible fMRI signals. However, the gadolinium
compounds used in humans have very short blood half-lives, and no suitable
agents are approved at this time.
For human studies, the most widely utilized technique is based upon
changes in an intrinsic contrast agent, deoxygenated hemoglobin. Its develop-
ment followed from the classic work of Pauling and Coryell (197,198) on the
diamagnetic versus paramagnetic state of oxyhemoglobin and deoxyhemoglo-
bin, respectively. Subsequent work by Thulborn and colleagues (249) evaluated
the in-vitro effect of oxygenation on the MRI signal. Independent groups led by
Ogawa and Turner extended these observations to note similar changes alter
T
2
-
weighted signals in vivo in mammals. Parallel work by Detre and colleagues
(77) demonstrated how to use
T
1
-weighted signals to quantify perfusion. With
important modifications, Kwong and colleagues (143) applied these develop-
ments to image oxygenation and flow changes associated with neural activity.
The work of Kwong and colleagues was first presented to other scientists at the
10th Annual Meeting of the Society of Magnetic Resonance (August 1991) and
were rapidly replicated and extended (10,143,190).
