Signal Collection

Scalp EEG electrodes are pasted or glued to the scalp using the International 10-20 System of electrode placement. This system uses the distances between bony landmarks on the head to create a system of lines. Recording electrodes are then placed at intervals of 10 or 20 percent of the total length of these lines. The primary advantage of using such a proportional system is that it will identify the same relative position on the scalp regardless of head size. The standard EEG setup for adults consists of 21 recording electrodes plus 1 ground electrode, and electrodes for a single channel of EKG and to record eye movements. Each electrode position is identified by a letter and number. The letters indicate the position of the electrode on the head: Fp, frontopolar; F, frontal; C, central; T, temporal; P, parietal; O, occipital. Odd numbers are used over the left hemisphere and even numbers over the right hemisphere. A lowercase “z” indicates a midline scalp position. Additional scalp electrodes may be placed at smaller proportional distances within the 10-20 System to more precisely represent the electrical activity within a certain brain region. For example, the best localization to detect an anterior temporal spike in a patient is often not ideally covered by the 10-20 System arrangement, and special anterior temporal electrodes (T1 and T2) can be added.

Amplification and Filtering

All EEG activity is recorded with differential amplifiers. These devices measure the electrical activity at one electrode relative to another, thus eliminating much of the common activity between the electrodes (“common mode rejection”). Because artifact, both biologic and ambient, is relatively similar around the head, it will often be substantially eliminated through the use of differential amplifiers. This “cancelling out” of signal leaves the brain activity of interest and can notably improve the signal-to-noise ratio in the recording. The amplifier also increases the voltage difference, so it may be visualized. EEG sensitivity is the ratio of the input voltage to the signal deflection. It is measured in microvolts per millimeters. The commonly used sensitivity is 7 μV/mm but can be adjusted up or down to make the EEG easier to visualize.

Filters are used to minimize activity of relatively high or low frequency so that the waveforms in the most important range (1–30 Hz) can be recorded clearly and with minimal distortion. At present, EEG machines have three types of filtering. There is a low-frequency filter that removes the amplitude of slow waves, a high-frequency filter that reduces the amplitude of fast waves, and a notch filter that selectively reduces the amplitude of waves in a narrow frequency to remove electrical line interference. In North America, the notch filter is set at 60 Hz. At the current time, these filters tend to be created by analog technology. As a result, the filters are not absolute, and they do not perfectly remove or preserve all frequencies above or below the individual formal settings but instead provide a continuum of gradual filtering; consequently, they cannot get rid of all recorded artifacts and may, in fact, distort them to the extent that pathologic wave forms are no longer recognizable.

EEG Display

After collection, EEG signal is displayed on the screen in specific montages, or arrangements. As a general rule, modern montages allow for easy visualization of comparable scalp areas, so they may be assessed for symmetry. There are two primary types of display montages: bipolar and monopolar/referential. Bipolar montages consist of chains of electrodes, each one connected to one or two neighboring electrodes. The bipolar longitudinal pattern, also called the “double banana,” is a commonly used bipolar montage. It consists of a display in which each channel connects adjacent electrodes from anterior to posterior in two lines, essentially covering the parasagittal and temporal areas bilaterally. The midline electrodes are also linked in a chain fashion. The bipolar transverse montage links adjacent electrodes in a chain, going from left to right. Monopolar, or referential montages, connect each electrode to a single referential point. This reference can be either another electrode on the scalp or a mathematical combination of signals, such as a mathematical average reference.

Localization of abnormalities in a bipolar recording system involves identifying the head region with the phase-reversal and assuming that the abnormal signal was generated within this head region. Localization of abnormalities in a referential system involves identifying the head region with the highest amplitude abnormalities and assuming that the abnormal signal was generated within this head region.

The major disadvantage of bipolar signal collection is that there can be in-phase cancellation of biological activity. In other words, if the biologic waveform at the two points compared is relatively synchronous, with respect to both time and amplitude, the differential amplifier can “cancel” them out, which can lead to false localization of low-amplitude phenomenon. The strength of referential montages is that in-phase cancellation does not occur. However, there is no perfect reference; the disadvantage of monopolar displays is the chance that the reference will be contaminated with signal, making EEG interpretation more difficult. In EEG interpretation, multiple montages should be used, and all abnormalities should be confirmed on multiple montages to determine that they are pathologic not simply a reflection of the method of signal display.