D1 Electroencephalography (EEG)

Topic

The electroencephalogram (EEG) signal recorded at the scalp originates from the electric field generated by the synchronous activity of large populations of cortical neurons. Within these populations, it is primarily the postsynaptic potentials summed within cortical columns that generate this electric field; this field propagates from the cortical columns to the scalp electrode according to a volume conduction model. The more synchronous the activity of the cortical neurons, the more intense the EEG signal ultimately recorded by the electrode.

Once recorded, the EEG signal can be decomposed in the frequency domain using its power spectral density, which quantifies the amount of EEG signal activity corresponding to each classic frequency band: delta, theta, alpha, beta, and gamma. The more power the EEG signal concentrates in a specific frequency band, the greater that band's contribution to the total recorded activity.

Based on the same EEG signal recorded at the scalp, the inverse source localization problem seeks to reconstruct the distribution of cortical activity that best explains the voltage pattern of the EEG signal; this is typically achieved through regularized minimization methods that identify—among all possible cortical activity distributions—the one that provides the best fit to the observed EEG signal.

When recorded simultaneously across multiple electrodes, the EEG signal also allows for the calculation of inter-channel coherence, a measure quantifying the functional coupling between the distinct cortical regions generating the signal at each channel. The higher the coherence between EEG channels, the stronger the functional coupling between the corresponding cortical regions. Finally, the electroencephalogram signal enables the automatic detection of epileptiform discharges, applying the same amplitude and duration criteria previously established to characterize synchronized discharges in the epilepsy model. Overall, it is the electroencephalogram signal—spanning everything from the electric field generated by cortical neurons to frequency analysis, source localization, and inter-channel coherence—that allows for both the characterization of normal brain activity and the detection of pathological patterns such as epileptiform discharges.

ID:4297