Biomedical Engineering Reference
In-Depth Information
Stimulus Evoked Cortical Activity
foot shock
foot shock
6.0
5.0
4.0
3.0
2.0
1.0
0.0
1 3 5 7 9 11131517192123252729313335373941
−
1.0
Acquisition Periods (6 sec)
Activational Maps of Primary Somatosensory Cortex
FIGURE 16.52
Functional imaging with fast spin echo. Images show percent change in BOLD signal in the
right somatosensory cortex (boxed area in activational maps) in response to electrical stimuli to the left hind limb
paw of a rat.
From [5], with permission from Elsevier.
with spatial sampling. Unlike fMRI or PET images that provide indirect or delayed clues to
synaptic events through changes in the metabolism or blood flow, MEG locates regions of
the brain that respond directly to stimuli and yields corresponding time responses at vari-
ous spatial locations.
To accomplish these tasks, a subject is placed into a helmet-like enclosure that houses a
semispherical array of hundreds of magnetic sensing coils. Because the brain's dynamic
magnetic field is weaker by five orders of magnitude than magnetic background noise, spe-
cial noise-suppressing coil configurations, as well as ultrasensitive detectors, superconduct-
ing quantum interference devices (SQUIDs), and shielded chambers are needed. The signals
from the coils as well as their locations on the semispherical grid are employed in solving
the inverse problem of determining the source location of the signals in the brain. These cal-
culations are repeated for different combinations of signals until active regions of the brain
are identified with their associated time waveforms. These waveforms are combined with
MRI and/or fMRI imaging to aid in localization. A resulting MEG neuroimage is shown
in Figure 16.54. In this experiment, parallel localization (“where”) and recognition (“what”)
