Biomedical Engineering Reference
In-Depth Information
fMRI allows the experimenter to determine which parts of the brain are
activated by different types of sensory stimulation, motor activity, or cogni-
tive activity. For instance, fMRI can be used to study responses related to
visual or auditory stimulation, the movement of a subject's fingers, or the
imagined rotation of 3D objects.
The subject in an fMRI experiment will lie within the magnet, and a partic-
ular form of stimulation is applied or task performed. For example, the subject
may wear special glasses so that pictures can be shown during the experiment.
Then, MR images of the subject's brain are taken, starting with a single high
resolution scan. This is used later as an anatomical substrate for overlaying
the brain areas which were activated by the stimulus. Next, a series of low res-
olution scans (the raw functional images) are taken, one every few seconds;
normally, 100 or more such scans are obtained. During some of these scans,
the stimulus (in this case the moving picture) will be presented, and during
others the stimulus will be absent. These images are sensitive to changes in
blood flow and
or blood oxygenation in the brain caused by brain activity.
The fMRI images taken during activation can be compared with those taken
during rest in order to see which parts of the brain were activated by the stim-
ulus or activity.
Other functional neuroimaging methods exist, such as PET (positron emis-
sion tomography), EEG (electroencephalography), and MEG (magnetoen-
cephalography). These differ from fMRI in their temporal and spatial
resolutions as well as the type of physiological response measured. For
instance, fMRI measures local blood oxygenation changes, PET measures
either blood flow or metabolic activity (a more direct measure of activity
than blood flow), EEG measures induced electrical signals on the scalp, and
MEG measures induced electrical currents within the cortex. EEG and MEG
have very high temporal resolution (milliseconds) but poor spatial resolu-
tion (centimeters); PET has poor temporal resolution (tens of seconds) and
intermediate spatial resolution (many millimeters), while fMRI has an inter-
mediate temporal resolution (seconds) and good spatial resolution (millime-
ters). PET also requires that a radioactive agent be injected into the subject,
while the other methods are relatively non-invasive.
8.1.1
BOLD Contrast and Brain Function
Activity in a certain brain area causes an increase in both local blood flow rate
and relative proportion of oxyhemoglobin to deoxyhemoglobin (in local
blood vessels). A T2-weighted MR sequence is sensitive to this change in
blood oxygenation; this is known as the BOLD (blood oxygenation level
dependent) effect
9,12
and is dependent on field strengths as well as physiolog-
ical and other factors. Therefore, as blood flow and oxygenation levels
increase in metabolically active regions, the MR signal increases; this is used
to infer neuronal activity. It is important to use an MR method which can
acquire T2-weighted images very rapidly. The most commonly used method
