5 Electroencephalography

What is Electroencephalogram (EEG)?

An Electroencephalogram (EEG) is an effective modality which helps to detect abnormalities within brain waves, or in the electrical activity of the brain. An EEG functions to acquire real-time signals which corresponds to a section of the scalps surface area, in order to measure the recordings of electrical signals. Before the testing process, electrodes are placed on the scalp via conductive gel, following the international 10-20 system. Electrodes are circular metal disks with thin wires that can detect electrical signals, which result from the activity of the brain cells. For the duration of the test, the technician will present verbal instructions to follow which are intended to stimulate the brain in different ways. For instance, the verbal instructions can consist of simple hand and eye movements in order to detect signals.

EEGs are a continuous measure of electrical brain activity, used to detect brain abnormalities, which is presented in the form of a recording. Often times EEG recordings and event related potentials (ERPs) are categorized as the same thing. D. Brandeis & D. Lehmann(2012) describe ERPs as “a recording of the electric field which the brain produces in fixed time-relation to an event”. As ERPs open what we call a time and space window onto covert steps of brain information by processing the overt behavior or private experiences. ERPs are classified as a non-invasive method which resolves the dynamic pattern of events in the human brain, down to the millisecond range D. Brandeis & D. Lehmann (2012). EPRs are simply small changes in the scalp-recorded EEG which are time-locked to the onset of an event. As an EEG provides an excellent medium to understand neurobiological dysregulation, with the potential to evaluate neurotransmission. Overall, this helps to capture neural activity related to both sensory and cognitive processes of the brain S. Sur & K. Sinha (2009). 

Electrode Placement and Labelling

The International 10-20 system is an internationally recognized system that is used to both describe and apply the location of electrodes to the scalp, for EEG testing the standardized purpose is to ensure all results can be complied for analysis. This system is strategically based on the relationship between the location of the electrode placement site and the underlying area of the brain Rojas and Colleagues (2018). The “10” and “20” of the system refers to the actual distance (10% or 20%) between the adjacent electrodes, to the distance of the skull. Each electrode placement corresponds with a letter in order to identify the lobe or area of the brain that is being read Rojas and Colleagues (2018). Along with each identifying letter there is also a number that corresponds, by following a specific system. Even-numbered electrodes refer to the electrodes that are placed on the right hemisphere of the head whereas, odd-numbered electrodes refer to the electrodes that are placed on the left hemisphere of the head Rojas and Colleagues (2018).  These specific anatomical landmarks are used not only for essential measuring, but also the positioning of  each electrode.

What are EEGs most commonly known to detect?

As previously mentioned, an EEG is a neurological test which can help to detect changes in brain activity that might be useful in the diagnosis of brain disorders. EEGs are meant to provide results at the time of the test but cannot provide results for what happens to a brain at any other time after the test occurs. Clinically, EEGs are most commonly used in the process of diagnosing epilepsy or other seizure related disorders. As well, EEGs are also performed in other studies such as sleep studies and monitoring the depth of anesthesia studies, as this presents the opportunity to monitor changes in human brain electrical activity of consciousness. One common misconception with EEGs is that the test itself only gives information about electrical activity of the brain. In other words, an EEG cannot provide any further evidence of damages or physical abnormalities with the brain. This is what other modalities such as magnetic resonance imaging (MRI) and computed tomography (CT) scans are used for, just to name a few.

How to extract data from an EEG?

The traditional method of extracting data from an EEG has many features. Often times after an EEG has taken place, the main focus is to extract the frequency patterns that were collected from the broadband EEG. Here there are several software packages which can be used for the analysis of the data. Often times the software will allow the change of frequency bands based on the routine definitions (see figure 3.1. For instance, Delta 1-4Hz). Related to Wei and colleagues (2019) work, the discussion of epilepsy as a neurological disorder is present, as clinicians usually diagnose epilepsy by interpreting EEG manually. This proposes a novel automatic epileptic EEG detection method based on convolutional neural network (CNN), with two innovative improvements, treating this task as a big data classification issue. Wei and colleagues (2019) bring forward the idea that an expert understanding of neural activity and frequency patterns must be obtained in order for a timed-series EEG recording to be extracted.  This process functions by a sliding window, which then passes through the CNN model and acts as an epileptic EEG classification model. Results achieve high sensitivity and specificity which yield novel improvements in the automatic epileptic EEG classification performance. In short, Wei and colleagues (2019), propose that extraction follows a pattern recognition approach that discriminates EEG signals recorded during different cognitive conditions.