Biomedical Engineering Reference
In-Depth Information
There are several scales at which bioelectricity can be described and measured.
At a microscopic scale, microelectrodes are placed inside or in very close vicinity
to neurons (see Chap. 8): at this scale, it is possible to measure APs as well as LFPs
which correspond to the local electric potential within the cortex as generated by the
nearby neurons. At the mesoscopic scale (below a square millimeter), intracortical
recordings (stereotaxic EEG, SEEG) only mesure the local field potential. Non-
invasive measurements of the electric potential via EEG or the magnetic field via
MEG are made on the scalp, and the spatial extent of brain activity to which these
measurements can be related does not yet meet a consensus, but lies between a
square millimeter and a square centimeter. Similarly to SEEG, MEG and EEG
mostly see PSPs: this results both from the faster decrease of the AP field with
distance and from the summation argument used for LFPs.
The remainder of this chapter mainly focusses on SEEG, MEG and EEG and
thus on PSPs.
7.1.2
Measuring Brain Activity
Several well-established techniques allow to non-invasively probe the brain in
function. Some are sensitive to the metabolic activity of the brain, and others to its
electric activity. Functional Magnetic Resonance Imaging (fMRI), which is sensitive
to the ratio between oxygenated and de-oxygenated haemoglobin, is a metabolic
imaging modality. It is an established technique for mapping regions involved in
cognitive tasks. Because of its spatial encoding, one can estimate from fMRI the
activity within the brain volume, with millimeter resolution. Unfortunately, the
recorded signal, and thereby the detection of active areas, relies on the slow hemo-
dynamic response to neuronal activity (taking several seconds to reach its peak).
Therefore, the temporal sequence of activated regions is difficult to estimate through
this technique. In contrast, electroencephalography (EEG) is a more direct reflection
of neuronal function. It records variations of electric potential with millisecond
resolution, at the time scale of neuronal synaptic activity. The EEG measures
variations of electric potential on the scalp, produced by the pyramidal neurons
of the gray matter. These brain sources actually produce an electromagnetic field,
the magnetic component of which can be measured by magnetoencephalography
(MEG). MEG and EEG thus measure two complementary consequences of the same
brain sources.
Localizing active brain areas from EEG requires to solve a difficult inverse
problem [ 1 ]. The resulting spatial resolution within the brain is of the order of
a centimeter for EEG and MEG. A synopsis of most current brain exploration
techniques is presented in Fig. 7.1 , allowing to compare their respective time
and spatial resolutions, and their degree of invasiveness. MEG, EEG and fMRI
rank among the least invasive techniques. Local field potentials are measured by
intra-cerebral electrodes: because their recording sites are only a few millimeters
away from the sources of activity, they do not suffer from convolution effects
Search WWH ::




Custom Search