Artifacts During Short-Term Interictal Electroencephalographic Recording in Dogs
ABSTRACT
The electroencephalogram (EEG) is an electrodiagnostic technique widely used in both scientific research and clinical medicine. It makes it possible to study the neurophysiology of brain activity by recording real-time changes in electrical potential produced by cortical activation. The importance of EEG in diagnosing canine epilepsy demonstrates its usefulness when the owner's description of crises is not clear or when the episodes cannot be differentiated from behavioral or cardiac disorders. However, EEG recordings also often record electrical activity from sources other than the brain, which may interfere with the clinical event-related signal. This activity is known as artifactual electrical activity, and the signal changes recorded in these cases corrupt the trace when they are superimposed on brain activity or even, in some cases, mimic pathologic abnormalities. The first step in analyzing and interpreting EEG traces is to recognize artifacts and other specific transient events. This retrospective study set out to ascertain whether artifacts comparable to those described in humans are observed in protocols used to perform short-term interictal EEG for canine epilepsy and how these can be classified.
Introduction
Electroencephalography is an electrodiagnostic technique widely used in both scientific research and clinical medicine.1 It enables study of the neurophysiology of brain activity by recording real-time changes in electrical potential produced by cortical activation.1,2 These events can be recorded more or less invasively depending on the electroencephalogram (EEG) technique, either directly on the cortex or on the scalp using surface or needle electrodes, and also over a variable duration.
Over the years, many studies in veterinary medicine have been published on the different EEG patterns that can be observed in patients suffering from primary (idiopathic or genetic), secondary (metabolic/structural or symptomatic), and unknown (cryptogenic) epilepsy. The latter category includes forms associated with meningoencephalitis, intoxication, and also forms deemed likely to be only symptomatic.1,3–8
Several authors have highlighted the importance of EEG in diagnosing canine epilepsy, and this is helpful when the owner's description of crises is not clear, or when the episodes cannot be distinguished from behavioral or cardiac disorders (e.g., syncope).6,9–12
In human medicine, EEG became more widespread during the second half of the last century, when it was widely used as a tool for diagnosing and locating intracranial disease. However, the advent of modern diagnostic imaging techniques such as computerized tomography and both classical and functional MRI has greatly reduced its sphere of use. Indeed, these diagnostic techniques make it possible to visualize intracranial structures with greater spatial resolution than functional diagnostic techniques (such as EEG), proving to be considerably more sensitive and specific in the neurolocalization of structural diseases underlying secondary epilepsy. Primary (idiopathic) or cryptogenic epilepsy can only be assumed when no macroscopic structural changes of the brain parenchyma can be found using these diagnostic techniques; this is because these forms of epilepsy only involve functional changes of the brain parenchyma without structural abnormalities.13
The information provided by EEG is purely functional, with high temporal resolution on a millisecond scale making it possible to view even transient abnormalities, such as those characterizing paroxysmal changes in cortical activity. This is the reason for the recent renewed interest in this examination technique during the study of epileptic disorders in both humans and animals.8,14,15
EEG traces in approximately 20–50% of human patients suffering from epileptic syndrome show altered interictal paroxysmal patterns as early as the first recording. In the same way, it seems that this activity can also be recorded in 20–86% of canine patients suffering from epileptic syndrome.3,5,10,16
Short-term interictal EEG (STI-EEG) is a noninvasive technique, lasts around 20 min, and is performed during the interictal period (i.e., during the interval between two epileptic seizures). It differs from both video EEG monitoring and ambulatory EEG, which involve much longer observation periods and are aimed at capturing ictal events. All types of EEG record the electrical activity generated by a group of cortical pyramidal neurons. They also provide direct information about cortical pyramidal neuron function as well as indirect information about neurons deeply located inside the brain, such as the nuclei of the thalamus or brainstem, which act as modulators of cortical activity through synchronization and desynchronization mechanisms.2
However, all types of EEG recordings can also record electrical activity from sources other than the brain, interfering with the clinical event-related signal. This activity is known as artifactual electrical activity, and the signal changes recorded in these cases corrupt the trace when they are superimposed on brain activity or even, in some cases, mimic pathologic abnormalities.17 The first stage in analyzing and interpreting EEG traces is to recognize artifacts and other specific transient events.18
The frequency with which EEG artifacts occur can be calculated if necessary, but cannot be scientifically predicted, either in practice or in theory. It depends on the large number of variables involved in procedures, the environment in which the study is carried out, and the patient's anatomic characteristics. However, if the main causes of EEG signal disturbance and the associated waveforms is known, we can minimize their effects during the test or, at least, correctly interpret the traces obtained.
The purpose of this retrospective study was to ascertain whether artifacts comparable to those described in humans are observed in protocols used to perform STI-EEG for canine epilepsy and how these can be classified according to the standardized computer-based organized reporting of EEG (SCORE).19
Materials and Methods
This is a retrospective study based on the analysis and review of raw standard STI-EEG traces recorded at the Associazione Professionale Neurovet between 2011 and 2013. The recordings and retrospective surveys were conducted by the authors with the cooperation of the Department of Veterinary Science staff of the University of Parma.
Samples
The trace samples used for the study comprised 200 randomly selected records obtained from subjects with varying prosencephalic disorders, investigated between 2011 and 2013. Every patient included in this study had epileptic seizures as the primary complaint, while other prosencephalic signs, such as head pressing or altered mental state, were inconstant. The neurological diagnostic workout included MRI and cerebrospinal fluid examinations.
Randomization was carried out by means of computer generation of random number sequences to obtain a uniform distribution. The patient population was composed of 113 male and 87 female dogs of different breeds, weighing from 1.5 to 65 kg. The age varied from 5 mo to 16 yr, with a median of 3.9 years. Sixty patients were 1 yr old or younger.
Techniques
The STI-EEG recordings were made using digital equipment, and the test protocol required the use of 12 monopolar transcutaneous needle electrodes, one grounding electrode, and one reference electrode arranged in the configuration used by Pellegrino, except that the temporal electrode positions were changed to correct some impedance abnormalities (Figure 1).20
Citation: Journal of the American Animal Hospital Association 53, 2; 10.5326/JAAHA-MS-6486
Each recording lasted approximately 20 min, and each record was saved digitally using management software, so as to store the raw EEG data in individual filesa.
The anesthesiological protocol used to sedate patients consisted of a low-dose premedication (dexmedetomidine hydrochlorideb at 2 μg/kg) and intravenous sedation with propofolc as a continuous pump infusion (loading dose of 50 mg/kg/h), which could be adjusted as necessary. Optimal anesthesia required that the patient's palpebral reflex was still present while voluntary movements and the swallowing reflex were depressed.18,21
Data Analysis
An operator manually analyzed raw (unfiltered) traces, scrolling through records in order to display them, because dedicated software for the automated identification of the artifacts was not available. Only an alternating current filter was utilized to remove artifacts. Both bipolar and monopolar montages were used to study all traces.
It was possible to identify several types of artifact on each trace and determine whether or not these might hinder satisfactory final analysis.
Artifacts were classified based on information provided by the above-mentioned SCORE and characterized based on information taken from the main bibliographical reference sources for human EEG.19
Signal changes were divided into two main categories: physiological and non-physiological. The former are produced by biological activities originating from the patient, while the latter are produced by electromagnetic field disturbances caused by the equipment used (hardware) and by other sources present in the environment where the EEG is performed.
The palpebral reflex (blink) artifact generated by the operator was not included in the calculation because it was often deliberately provoked to test the depth of anesthesia. However, it was described because it can occur autonomously and therefore could be confused with a transient paroxysmal event.
While our observations are purely retrospective and do not have predictive value owing to the particular conditions under which each study took place, the recorded frequencies may provide useful information that in principle can be adapted to the various situations in which operators work.
Results
During EEG analysis, various artifactual track patterns were observed, identified, and divided in the two SCORE categories: physiological and non-physiological artifacts. There was a complete match between the kinds of artifacts identified in our study and the SCORE classification. The classified artifacts are summarized in Table 1. The most frequent artifacts visualized are shown in Figures 2–10.
Citation: Journal of the American Animal Hospital Association 53, 2; 10.5326/JAAHA-MS-6486
The artifacts identified are listed in sections 3.1 and 3.2 below, together with a description of their causes, information useful for identifying them, and the procedures to be adopted to minimize their effects. Any similarities or differences with human subjects are also included.
Physiological Artifacts
Eye Movements
This artifact was the most common and was recorded in 176 patients (88%). Eyeball movements generated visible artifacts on traces from frontopolar (Fp) electrodes and sometimes also those from frontal (F) electrodes. The trace showed a mono- or biphasic sharp waveform or a complex polyphasic slow waveform (1 Hz). Some main features are shown in Figure 2. Upward movements of the eyeball caused the cornea (positive pole) to shift towards the frontopolar electrode and were typically recordable as monophasic positive (downward) deflections. The opposite occurred with downward movements (negative upward deflection). The duration of deflection reflected the duration of movement. The recorded potentials had amplitudes of 50–100 μV and were bilateral, their intensity being inversely proportional to the distance between the electrode and eye.
Horizontal movements mainly concerned the ipsilateral frontal electrodes. Similar to vertical movement, the cornea deviated towards the ipsilateral electrode during horizontal movement, causing a positive deflection, and away from the contralateral electrode, causing a negative deflection. Maximum deflection amplitude was obtained in the ipsilateral electrode.
Eye Blinks
Eye blinks were not included in the calculation because these are often deliberately provoked by the operator to test the depth of anesthesia (Figure 2A).
A blink causes the positive pole (i.e., cornea) to move closer to the frontopolar (Fp1-Fp2) electrodes, producing symmetric downward (negative) deflections. Unlike movements of the eyeball, the potentials generated by blinks were rapid, and may be mistaken for epileptiform activity.
EMG
Electromyographic (EMG) activity was one of the most common and major sources of interference with EEG recordings. We recorded this artifact in 173 subjects (76.5%). In most cases, EMG activity was superimposed on the EEG and obscured concomitant EEG signals owing to its high amplitude and frequency. However, potentials of cortical origin expressed frequency and amplitude that were inversely proportional to one another (Figure 3). This artifact typically took the form of a sudden, arrhythmic “burst,” which appeared simultaneous with the start of muscle activity and was often macroscopically invisible. Prolonged duration and frequency of discharge (20–100 Hz, in most cases 30–50 Hz) was also helpful in identifying them.
Citation: Journal of the American Animal Hospital Association 53, 2; 10.5326/JAAHA-MS-6486
ECG Artifacts
Cardiac activity was recorded in 43 patients (21.5%), 28 of which were large breed dogs. The heart produced two types of EEG artifact: electrical and mechanical. Both types were rhythmic, normally regular, and were temporally correlated with heart contraction and simultaneous with electrocardiogram (ECG) activity.
Electrical artifacts showed only the QRS complex but not P and T waves. This type of artifact was more evident with referential montages, even when the ground electrode was not correctly positioned on the dog's neck. Moreover, it was not present simultaneously on all traces, particularly in longitudinal bipolar montages, and it was sometimes evident throughout the duration of the whole recording, while in some rare cases intermittent and undulating (Figure 4).
Citation: Journal of the American Animal Hospital Association 53, 2; 10.5326/JAAHA-MS-6486
Mechanical artifacts linked with cardiac activity were due to arterial pulse and may be considered “pulse” artifacts. Indeed, they were observed in those electrodes placed close to a vessel that produced a pulse wave determining the movement of the needle electrode, even if imperceptibly. A rhythmic slow-wave was seen, following around 100–300 ms behind the ECG artifact, depending on the size of the dog. This distortion occurred anywhere on the scalp and at times showed a more or less pointed form.
Ear Artifacts
Ear artifacts were recorded in 21 tracks (10.5%) and were obvious in relation to temporal (T) electrodes. This artifact appeared similar in morphology and amplitude on the trace to those produced by ocular motor activity (Figure 5).
Citation: Journal of the American Animal Hospital Association 53, 2; 10.5326/JAAHA-MS-6486
Non-Physiological Artifacts
Non-physiological artifacts were signal distortions caused by sources external to the patient. The equipment used (hardware) and environment in which the study was performed were the two main sources of these artifacts.
Alternating Current Artifacts (50–60 Hz)
Very high frequency (50–60 Hz or above), long duration, or alternating current generated continuous rhythmic changes in electric potential in 160 (80%) of the 200 tracks (Figure 6). The resulting artifact appeared as a series of sharp, rapid, positive–negative deflections in one or more electrodes.
Citation: Journal of the American Animal Hospital Association 53, 2; 10.5326/JAAHA-MS-6486
Electrode Artifacts
This was not a common artifact, and it was detected in 15 (7.5%) of our tracks. This type of artifact occurred in one of two ways. In the first case, brief transients limited to one electrode were observed, while in the second case, low-frequency, high-amplitude rhythms were observed across the entire scalp. The morphology of these EEG signal changes was totally abnormal in brain waves, making it easier to identify the artifact (Figure 7).
Citation: Journal of the American Animal Hospital Association 53, 2; 10.5326/JAAHA-MS-6486
Induction Artifacts and External Equipment Artifacts
These signal distortions were the most common artifacts (24 tracks, 12%) caused by equipment adjacent to the patient, such as heating blankets, neon lights, pulse oximeters, infusion pumps, and various types of physiological monitors typically present in intensive care units where the EEG was performed. In this case, changes in EEG signal consisted of high-frequency, high-amplitude waves persisting for as long as the equipment was working (Figure 8). A further source of interference was a mobile telephone, as the incoming and outgoing signals created EEG artifacts, with positive vertical spike trains being seen, even in just one electrode; this was similar to the artifact caused by the medical equipment (as shown in Figure 8). Removing the source of the signal distortion extinguished the artifact.
Citation: Journal of the American Animal Hospital Association 53, 2; 10.5326/JAAHA-MS-6486
Whenever the environment in which the procedures were carried out was inadequately grounded (admissions, kennels), the electric motors connected to the network generated electrical interference compromising the recordings and, rarely, the EEG equipment itself (Figure 9).
Citation: Journal of the American Animal Hospital Association 53, 2; 10.5326/JAAHA-MS-6486
Citation: Journal of the American Animal Hospital Association 53, 2; 10.5326/JAAHA-MS-6486
Salt Bridge Artifacts
This artifact typically appeared in two adjacent recording channels when two electrodes were too close to each other. Waves recorded in this way in the unipolar montage appeared almost superimposable, while an isoelectric wave was recorded in the longitudinal montage, making it easier to identify the anomaly (Figure 10). There were 11 (5.5%) salt bridge artifacts in our study, 9 of which belonged to EEG traces of small patients.
Discussion
During EEG acquisition, the type and frequency of artifacts can be calculated, if necessary, but cannot be scientifically predicted. They depend on the many variables involved in the procedures used, the environment in which the EEG is performed, and the characteristics of each individual patient. However, knowing the main sources of interference with the recording and the associated waveforms usually makes it possible to minimize the signal changes or, if this is not possible, to correctly interpret the traces obtained.2
Based on our results, it can be stated that artifacts commonly observed in human EEGs also occur in canine EEGs.
Physiological Artifacts
Eyeball Movement
Eyeball movement in three orthogonal planes directly determined the wave type recorded on the trace, as described in previous papers. Indeed, the eye is a dipole with a positive pole at the cornea and a negative pole at the retina. Its movement causes a change in the surrounding high-amplitude (if compared with that generated by brain waves) electric field, which can be picked up by all the electrodes adjacent to the eye. The other sources of artifacts include the electrical potential produced by muscle cells (EMG) involved in contracting the orbital and extraorbital muscles.22,23,24
Eye Blink
EMG potentials from muscles in and around the orbit are the other source of ocular artifacts. Bilateral or unilateral blinks can be involuntary or may be provoked to assess the depth of anesthesia, as in this study (Figure 2A).2,22,24
EMG
Any movement or muscle contraction while the EEG was being performed generated artifacts, which were caused either by the electric field generated by muscle activity or by movement of the electrodes themselves. However, the frequency of EMG artifacts could not be calculated by visual analysis, because at normal sweep speed they appear on traces as a disorganized, continuous palisade. The hardware filters of the EEG equipment eliminate incoming signals in excess of 70 Hz; therefore, the signal loses its typical EMG appearance.22–24 EMG activity, usually produced by muscle fibrillations and twitches that are macroscopically invisible, may last from fractions of a second to the entire duration of the recording, similar to events in humans.23
EMG artifacts cause severe interference because they prevent the underlying EEG signals from being recognized and evaluated.2 Twitches are the most common EMG artifacts and reflect muscle action potential normally recorded during EMG. They mainly occur in the frontal and temporal muscles. The presence of muscle potential is reflected in the trace by thick solid spikes (in the form of a palisade) recorded on the electrodes concerned (normally more than one).22–24
Sometimes the artifact can be mitigated using appropriate software filters (50–70 Hz) installed in the devices by many manufacturers, although this may result in the loss of information; alternatively, mitigation may take place by means of an intramuscular injection of small doses of local anesthetics near needle electrodes (e.g., a 2% wt/volume lidocaine solution). Another solution favored by some authors is the intravenous injection of a neuromuscular blocking drug, such as rocuronium bromide dosed at 0.14 mg/kg, as described by Brauer.18 As part of the above-mentioned study, subjects were on intermittent positive-pressure ventilation and received ventilatory assistance during spontaneous recovery from neuromuscular blockade.24
ECG Artifacts
Cardiac activity produces another important artifact, as shown. Electrical ECG artifacts typically occur in the case of scalp ECG due to the distance involved and the poor amplification, and only the QRS complex is usually recorded (not P and T waves). Thus, the complex may be monophasic, biphasic, or triphasic on the EEG, and the amplitude may be in relation to the dog's body structure; for example, animals with a short neck in relation to chest size may manifest this high-amplitude artifact.22,23 This is a likely reason why, in our experience, cardiac artifacts are more common in large-breed dogs (28 among 43 patients). The artifact must be interpreted by noting its correspondence to the ECG channel, thus avoiding misinterpretation, even if its regularity is interrupted by cardiac arrhythmias (Figure 4).22–24
Mechanical artifacts linked with cardiac activity can be tested by gently touching the electrode with one finger, and, if confirmed, the electrode should be wiggled gently until the desired result is achieved. In some rare situations, the trace can also record the ECG activity of people in close contact with the patient during the test, even though this is a rare event and can easily be checked by removing the source of the interference.2
Non-Physiological Artifacts
Alternating Current
These potentials are generated directly by the energy source of the device used and are contingent upon both the electrical cable and the filter. Careful grounding of the patient often removes this artifact, as it is eliminated when the impedance between an electrode and “grounding” becomes sufficiently high (5,000 ohms or more). Indeed, if this situation occurs, the ground electrode becomes an “active” electrode, recording signals and generating an artifact. Otherwise, this problem can be partially or completely resolved by installing the correct software filter, as in our experience (Figure 6).2,25 Proper filters are currently included in all electromyography equipment on the market.
Electrode Artifacts
These individual or multiple sharp waves are regularly limited to a single electrode or, in the case of longitudinal montages, to all montages including this electrode. Whenever one or more electrodes are moved, it is recorded as an impedance in the electrodes involved that persists for the duration of the movement itself (Figure 7C). The sudden change in impedance causes a brief discharge, reflected by a “vertical” or very rapid transient and not by a gradual deflection, followed by a rapid return to the original potential (baseline; Figure 7A, B). Electrode popping does not affect background activity.2,25 If just one electrode is affected, the artifact is caused by a brief EEG discharge created between the electrode and the subcutaneous tissue. These events are known as electrode pops and reflect the capacity of the tissue and the electrode to form a capacitor, accumulating electric charges in the subcutaneous fluids.2
Induction Artifacts End External Equipment Artifacts
Induction artifacts and artifacts due to external equipment are sometimes present when the equipment is not directly connected to the patient, even by just one electrode.25 If this is not solved by moving the source of the artifact away, the only effective solution is to unplug the power supply cable of the responsible equipment.23
Salt Bridge Artifacts
The salt bridge artifact is related to the position of two electrodes whose recordings are the same. The closer the electrodes, the more similar the electromagnetic field potentials are to one another. This is the most likely reason why nine of the 11 patients on which this happened were small-breed dogs.24,25
In general, one good way to evaluate a change in signal involving a single electrode is to change its position and depth, checking the resulting changes in impedance. If the necessary response is not achieved, it is recommended to replace the malfunctioning electrode to check whether the problem is due to the electrode itself or the recording channel.2
The technical difficulties involved in performing EEG studies and in the ongoing quest for reliable, standardized recording and anesthesiology protocols are some of the reasons why EEG is rarely used in veterinary medicine. Indeed, the qualitative interpretation of traces is entirely up to the operator and requires a lengthy training period and solid theoretical grounding.8,18 Therefore, the essential first step is to be able to distinguish the valid EEG signals caused by the dog's brain activity from EEG artifacts.
Artifacts in dog EEGs are very similar to those that occur in human EEGs in terms of typology and frequency, so it is likely that automated or semi-automated techniques of artifact EEG detection and analysis can be used in the future. A semi-automated technique to control the quality of EEG signals is to delete the epochs altered by short-lasting artifacts. For instance, the technician responsible for this operation may delete the series of digitized samples immediately preceding an identified artifact.26
Eye movements and muscle potentials occur in most records of a few minutes' duration. They can distort power spectra and lead to the detection of transient nonstationarities that are difficult to distinguish from epileptiform events. Eye blinks and slow eye movements are bilaterally synchronous with a maximum in frontal derivations and represent an important contribution to the power in the delta band in these derivations. In the past, several methods have been proposed to avoid this type of artifact, involving subtracting the electro-oculogram (EOG). However, this approach may cause distortion of the EEG signals, as the EOG recording also contains brain signals.27–29
Another way to delete EOG activity from EEG traces can be achieved using a frequency domain approach. In fact, eye blinks and slow eye movements have different spectral properties and are transferred in different ways to the skull. Different types of gain functions for transferring both types of eye movements to the skull were computed.27,30–32
EMG signals affect the EEG power spectrum not only at very high frequencies (30–60 Hz) but sometimes also down to 14 Hz.33 Under normal conditions, there is very little EEG power in the 30- to 50-Hz band, so, if the power is significantly large, one must suspect contamination with EMG signals. A proposed technique to deal with this problem is the introduction of a reduction factor with which the activity in the beta band should be multiplied (for <1.5 μV/Hz, the reduction factor is 1; for >1.5 μV/Hz, the reduction factor decreases linearly to 0.1 as the spectral activity increases up to 5.0 μV/Hz).34
The need to avoid contamination with artifacts of relevant EEG features is so pressing that this area of EEG signal analysis has been constantly evolving. Recently elaborated methods are based on decomposing a set of EEG signals into components that should represent the artifact and the EEG signals, respectively. Several strategies and combinations of approaches have been compared, in particular with respect to their practical implementation.35
Conclusions
The main purpose of this retrospective study was to ascertain whether artifacts in canine STI-EEG were comparable to those described in human medicine, and this was achieved; artifacts commonly observed in human EEGs also occur in canine EEGs. Moreover, the EEG artifacts in dogs were easily classified according to the SCORE used. As mentioned in the Discussion, the essential first step in analyzing and interpreting EEG traces is to recognize artifacts and other specific transient events to be able to distinguish the valid EEG signals from EEG artifacts. These signal distortions can be corrected during the acquisition time or by the use of dedicated software, when it is available.
Although much remains to be learned in relation to this research topic, the authors hope that this work constitutes a step forward towards the wider application of this useful diagnostic technique.
Electrode arrangement for STI-EEG recordings. The montage is arranged as in the configuration used by Pellegrino, with the exception that the temporal electrodes are inserted subcutaneously, just rostral to the tragus and ventral to the zygomatic arch.20
(A) The bilateral vertical eye movement generates a positive (downward) slow peak in Fp2-F4 and Fp2-T4 for the left side, and Fp1-F3 and Fp1-T3 for the right side. It is associated with a sudden and rapid right-lateral contraction of the extraocular muscles due to the palpebral reflex (blink) elicited by the clinician at the 4th second of the epoch presented. (B) The biphasic positive to negative deflection in Fp1-F3 and Fp1-T3 is due to right eye oblique movement of the eyeball. This alteration is also appreciable, even if with lower amplitude, on the left side (Fp2-F4, Fp2-T4), on which there is a superimposition of a high-frequency low-amplitude artifact due to temporal-muscle contraction.
A mixed frequency electromyograph (EMG) artifact of prolonged duration on the right side (Fp1-F3; F3-P3; Fp1-T3) and on-coming on the left side in the second half of the epoch (Fp2-F4; F4-P4; Fp2-T4).
The electrocardiogram (ECG) artifact is visualized on all electrodes, with its maximum amplitude in the left-temporal electrode and Cz-Pz.
A bilateral and biphasic slow potential of 2 s duration, generated by ear movement, clearly visualized on parietal and temporal electrodes.
(A) A 60 Hz artifact from the alternating current that obscures EEG tracks in P3-O1 and T3-O1. (B) The same track is shown in Figure 5A after applying a 60 Hz filter.
(A) A single quick transient is visualized in all electrodes and is characterized by monophasic quick deflection of the potential recorded (“popping”). (B) The electrode impedance changes with 1 Hz frequency in Fp1 during the whole registration presented. (C) Poly-phasic haphazard transient in all the traces due to the accidental movement of the electrodes by the operator (approximately 2 s duration).
A multiple sharp positive deflection (4 s duration and 50–150 mV amplitude) characterises this epoch. It was caused by an electromagnetic source.
In all the recording electrodes, a rhythmic, abrupt onset of several high frequency and high amplitude poly-phasic deflections lapsed by lack of signal is shown.
The Fp1 and P3 electrodes are wrongly positioned (too close), therefore similar potentials are recorded to each other in the monopolar montage (A). The mistake is much easier to recognize in the bipolar montage (B) due to the “flattening” to the baseline in Fp1-F3.
Contributor Notes
