EEG Interpretation Research Articles

EEG in fitness to drive evaluations in people with epilepsy — Considerable variations across Europe

PurposeEpilepsy patients consider driving issues to be one of their most serious concerns. Ideally, decisions regarding fitness to drive should be based upon thorough evaluations by specialists in epilepsy care. In 2009, an EU directive was published aiming to harmonize evaluation practices within European countries, but, despite these recommendations, whether all epileptologists use the same criteria is unclear. We therefore conducted this study to investigate routine practices on how epileptologists at European epilepsy centers evaluate fitness to drive. MethodsA questionnaire was sent to 63 contact persons identified through the European Epi-Care and the E-pilepsy network. The questionnaire addressed how fitness-to-drive evaluations were conducted, the involvement of different professionals, the use and interpretation of EEG, and opinions on existing regulations and guidelines. ResultsThe questionnaire was completed by 35 participants (56 % response rate). Results showed considerable variation regarding test routines and the emphasis placed on the occurrence and extent of epileptiform discharges revealed by EEG. 82 % of the responders agreed that there was a need for more research on how to better evaluate fitness-to-drive in people with epilepsy, and 89 % agreed that regulations on fitness to drive evaluations should be internationally coordinated. ConclusionOur survey showed considerable variations among European epileptologists regarding use of EEG and how findings of EEG pathology should be assessed in fitness-to-drive evaluations. There is a clear need for more research on this issue and international guidelines on how such evaluations should be carried out would be of value.

Interpretation of electroencephalography in infants

Aim: Based on published reports and our own observations, we aimed to assign the graphical elements of EEG into normal and abnormal groups and then identify up to five main graphic elements in each group to facilitate visual analysis and interpretation of EEG in young children.Materials and methods. We searched for the relevant sources in the Medline and Medscape databases using the following keywords: «neonatal EEG», «neonatal seizures», «theta burst», «delta brushes», «trace discontinue», «burst-suppression», hypsarrhythmia», etc. Our own research was conducted using an Encephalan-EEGR-19/26 encephalograph equipped with children size gold cup electrodes with a diameter of 0.6 cm. Encephalograms were recorded from 10 electrodes according to the international “10-20” system.Results and discussion. In early childhood EEGs, two large groups of EEG graphic elements can be discerned: the likely normative graphic elements and patterns of pediatric EEG (normal patterns) and the likely non-normative (abnormal) graphic elements and patterns of pediatric EEG. In the likely normative group, the main features are represented by: theta bursts, delta brushes, the “intermittent EEG curve” pattern, the occipital theta rhythm, and slow-sleep waves. In the likely non-normative group, those are: paroxysmal EEG graphic elements, asynchronous EEG patterns, spike-wave discharges, 3 Hz peak waves, hypsarrhythmia, burst-suppression pattern, rolandic occipital spikes, and a slowing rhythm pattern.Conclusion. Along with the numerous attempts to characterize the age-dependent graphic elements at an early age, there are few reports concerning older children and adults. Here we try to overcome this discrepancy by identifying two large groups of graphic elements in EEG that are similar between infants and children of an older age. Such an approach may contribute to a better understanding of normal and pathological ontogenesis.

Ictal EEG source imaging and connectivity to localize the seizure onset zone in extratemporal lobe epilepsy

PurposeTo evaluate the yield of Functional Connectivity (FC) in addition to low-density ictal Electrical Source Imaging (ESI) in extratemporal lobe epilepsy (ETLE), using an automated algorithm for analysis. MethodLong-term EEG monitoring of consecutive ETLE patients who underwent surgery was reviewed by epileptologists, and seizure onsets characterized by rhythmical activity were identified. A spectrogram-based algorithm was developed to select objectively the parameters of ESI analysis. Two methods for SOZ localization were compared: i) ESI power, based on LORETA exclusively; ii) ESI + FC, including a Granger causality-based connectivity analysis. Results were determined at a sublobar level. The resection zone, in relation to 1-year follow-up surgical outcome, was considered as reference standard for diagnostic accuracy analyses. ResultsNinety-four seizures from 24 patients were analyzed. At seizure-level, ESI power showed 36 % sensitivity and 72 % specificity (accuracy: 45 %). ESI + FC significantly improved the accuracy, with 52 % sensitivity and 84 % specificity (accuracy: 61 %, p = 0.04). Results of ESI + FC were equally valuable in patients with lateralized or bilateral/generalized visual interpretation of ictal EEG. In a patient level sub-analysis, upon blinded clinical interpretation, ESI + FC showed a correct localization in 67 % of patients and substantial inter-rater agreement (kappa = 0.64), against 27 % achieved by ESI power, with fair inter-rater agreement (kappa = 0.37). ConclusionFC significantly improves SOZ localization compared to ESI solely in ETLE. Ictal ESI + FC could represent a novel option in the armamentarium of presurgical evaluation, aiding also in patients with visually non-localizable scalp ictal EEG. Prospective studies evaluating the clinical added value of automated low-density ictal ESI may be justified.

Practical Training of Anesthesia Clinicians in Electroencephalogram-Based Determination of Hypnotic Depth of General Anesthesia.

Electroencephalographic (EEG) brain monitoring during general anesthesia provides information on hypnotic depth. We hypothesized that anesthesia clinicians could be trained rapidly to recognize typical EEG waveforms occurring with volatile-based general anesthesia. This was a substudy of a trial testing the hypothesis that EEG-guided anesthesia prevents postoperative delirium. The intervention was a 35-minute training session, summarizing typical EEG changes with volatile-based anesthesia. Participants completed a preeducational test, underwent training, and completed a posteducational test. For each question, participants indicated whether the EEG was consistent with (1) wakefulness, (2) non-slow-wave anesthesia, (3) slow-wave anesthesia, or (4) burst suppression. They also indicated whether the processed EEG (pEEG) index was discordant with the EEG waveforms. Four clinicians, experienced in intraoperative EEG interpretation, independently evaluated the EEG waveforms, resolved disagreements, and provided reference answers. Ten questions were assessed in the preeducational test and 9 in the posteducational test. There were 71 participants; 13 had previous anesthetic-associated EEG interpretation training. After training, the 58 participants without prior training improved at identifying dominant EEG waveforms (median 60% with interquartile range [IQR], 50%-70% vs 78% with IQR, 67%-89%; difference: 18%; 95% confidence interval [CI], 8-27; P < .001). In contrast, there was no significant improvement following the training for the 13 participants who reported previous training (median 70% with IQR, 60%-80% vs 67% with IQR, 67%-78%; difference: -3%; 95% CI, -18 to 11; P = .88). The difference in the change between the pre- and posteducational session for the previously untrained versus previously trained was statistically significant (difference in medians: 21%; 95% CI, 2-28; P = .005). Clinicians without prior training also improved in identifying discordance between the pEEG index and the EEG waveform (median 60% with IQR, 40%-60% vs median 100% with IQR, 75%-100%; difference: 40%; 95% CI, 30-50; P < .001). Clinicians with prior training showed no significant improvement (median 60% with IQR, 60%-80% vs 75% with IQR, 75%-100%; difference: 15%; 95% CI, -16 to 46; P = .16). Regarding the identification of discordance, the difference in the change between the pre- and posteducational session for the previously untrained versus previously trained was statistically significant (difference in medians: 25%; 95% CI, 5-45; P = .012). A brief training session was associated with improvements in clinicians without prior EEG training in (1) identifying EEG waveforms corresponding to different hypnotic depths and (2) recognizing when the hypnotic depth suggested by the EEG was discordant with the pEEG index.

AEEG Use in Pediatric Critical Care-An Online Survey.

Background: Evidence supporting continuous EEG monitoring in pediatric intensive care is increasing, but continuous full-channel EEG is a scarce resource. Amplitude-integrated EEG (aEEG) monitors are broadly available in children's hospitals due to their use in neonatology and can easily be applied to older patients.Objective: The aim of this survey was to evaluate the use of amplitude-integrated EEG in German and Swiss pediatric intensive care units (PICUs).Design: An online survey was sent to German and Swiss PICUs that were identified via databases provided by the German Pediatric Association (DGKJ) and the Swiss Society of Intensive Care (SGI). The questionnaire contained 18 multiple choice questions including the PICU size and specialization, indications for aEEG use, perceived benefits from aEEG, and data storage.Main results: Forty-three (26%) PICUs filled out the questionnaire. Two thirds of all interviewed PICUs use aEEG in non-neonates. Main indications were neurological complications or disease and altered mental state. Features assessed were mostly seizures and side differences, less frequently height of amplitude and background pattern. Interpretation of raw EEG also played an important role. All interviewees would appreciate the establishment of reference values for toddlers and children.Conclusions: aEEG is used in a large proportion of the interviewed PICUs. The wide-spread use without validation of data generates the need for further evaluation of this technique and the establishment of reference values for non-neonates.

Rapid Response Electroencephalography for Urgent Evaluation of Patients in Community Hospital Intensive Care Practice.

Limited access to specialized technicians and trained neurologists results in delayed access to electroencephalography (EEG) and an accurate diagnosis of patients with critical neurological problems. This study evaluated the performance of Ceribell Rapid Response EEG System (RR-EEG), which promises fast EEG acquisition and interpretation without traditional technicians or EEG-trained specialists. The new technology was tested in a community hospital intensive care unit in Northern California. Three physicians (without previous training in EEG) were trained by the manufacturer of the RR-EEG and acquired EEG without the help of any EEG technicians. Time needed from order to EEG acquisition was noted. Quality of EEG and diagnostic information obtained with the new EEG technology were evaluated and compared with the same information from conventional clinical EEG system. Ten patients were tested with this new EEG technology, and 6 of these patients went on to have conventional EEGs when the EEG technicians arrived at the site. In these cases, the conventional EEG was significantly delayed (11.2 ± 3.6 hours) compared with RR-EEG (5.0 ± 2.4 minutes; P < .005). Use of RR-EEG helped clinicians rule out status epilepticus and prevent overtreatment in 4 of 10 cases. RR-EEG and conventional EEG systems yielded similar diagnostic information. RR-EEG can be set up by nurses, and diagnostic information about the presence or absence of seizures can be appreciated by nurses. The RR-EEG system, compared with the conventional EEG, did not require EEG technologists and enabled significantly faster access to diagnostic EEG information. This report confirms the ease of use and speed of acquisition and interpretation of EEG information at a community hospital setting using an RR-EEG device. This new technology has the potential to improve emergent clinical decision making and prevent overtreatment of patients in the intensive care unit setting while empowering nursing staff with useful diagnostic information in real time and at the bedside.

A novel approach based on wavelet analysis and arithmetic coding for automated detection and diagnosis of epileptic seizure in EEG signals using machine learning techniques

Epilepsy, a common neurological disorder, is generally detected by electroencephalogram (EEG) signals. Visual inspection and interpretation of EEGs is a slow, time consuming process that is vulnerable to error and subjective variability. Consequently, several efforts to develop automatic epileptic seizure detection and classification methods have been made. The present study proposes a novel computer aided diagnostic technique (CAD) based on the discrete wavelet transform (DWT) and arithmetic coding to differentiate epileptic seizure signals from normal (seizure-free) signals. The proposed CAD technique comprises three steps. The first step decomposes EEG signals into approximations and detail coefficients using DWT while discarding non-significant coefficients in view of threshold criteria; thus, limiting the number of significant wavelet coefficients. The second step converts significant wavelet coefficients to bit streams using arithmetic coding to compute the compression ratio. In the final step, the compression feature set is standardized, whereupon machine-learning classifiers detect seizure activity from seizure-free signals. We employed the widely used benchmark database from Bonn University to compare and validate the technique with results from prior approaches. The proposed method achieved a perfect classification performance (100% accuracy) for the detection of epileptic seizure activity from EEG data, using both linear and non-liner machine-learning classifiers. This CAD technique can thus be considered robust with an extraordinary detection capability that discriminates epileptic seizure activity from seizure-free and normal EEG activity with simple linear classifiers. The method has the potential for efficient application as an adjunct for the clinical diagnosis of epilepsy.

Developmental EEG hallmark or biological artifact? Glossokinetic artifact mimicking anterior slow dysrhythmia in two full term newborns

The aim of this paper is to describe an uncommon physiological EEG artifact in newborns caused by tongue movements (TM), mimicking anterior slow dysrhythmia (ASD). The subjects are two full-term newborns (39 weeks gestational age (GA)), admitted to the Neonatal Intensive Care Unit for respiratory distress. Both underwent polygraphic video-EEG recording in order to better characterize tremor-like movements of all four limbs that appeared 48 hours after birth. Multichannel video-EEG polygraphy was performed using the 10-20 electrode montage modified for neonates. Ninety minutes of EEG was recorded for each subject, capturing different behavioral states. Background EEG activity was normal for both subjects. During active sleep (AS), synchronous and symmetric slow activity was recorded over bifrontal head regions. For subject 1, bursts of monomorphic 2Hz delta waves, with an amplitude between 50-100μV lasting two seconds, were recorded and identified as anterior slow dysrhythmia. For subject 2, polymorphic 1-2Hz delta waves, 50-100μV in amplitude and lasting for 20 seconds, were recorded only during suction. After thorough analysis of simultaneous digital video recording synchronized with the EEG trace, this activity was thought to be compatible with glossokinetic artifact. Interpretation of neonatal EEG can be challenging; the background activity is frequently intermixed with physiological artifacts, such as ocular, muscle and movement artifacts, complicating the interpretation. Even continuous video-recording might not make the diagnosis immediately obvious. Therefore, when a rhythmic monomorphic pattern without evolution in amplitude or frequency is seen, we suggest that tongue movement artifact should be considered.

Buildout of Methodology for Meticulous Diagnosis of K-Complex in EEG for Aiding the Detection of Alzheimer’s by Artificial Intelligence

Application of artificial intelligence (AI) in health-care detection is a domain of exceptional research and interest in today’s world. And hence among this domain, a considerable inclination is toward creating a smart system that is AI for aiding identification of brain-related disease—Alzheimer’s—using electroencephalogram (EEG). Certain AI-based techniques as well as systems have been created for EEG examination and interpretation, but they have a common drawback that is lack of shrewdness and acuteness. Therefore, to overcome these drawbacks, a different methodology or technique is suggested in this paper which is able to mold the AI technique for better EEG Cz strip K-complex identification. This suggested method and structure of AI detection system is relied on quantitative scrutinization of Cz strip and embedding-established EEG explication principles for detection of K-complex and Alzheimer’s. This technique unconditionally relied on facts and information of neuroscience that are applied by expert in health care such as neurologist to create a detailed review of sick person’s EEG. The suggested technique also allots a potential of learning on its own to the AI so that it can apply the events in future examinations.

Bedside interpretation of simplified continuous EEG after cardiac arrest.

Continuous EEG-monitoring (cEEG) in the ICU is recommended to assess prognosis and detect seizures after cardiac arrest but implementation is often limited by the lack of EEG-technicians and experts. The aim of the study was to assess ICU physicians ability to perform preliminary interpretations of a simplified cEEG in the post cardiac arrest setting. Five ICU physicians received training in interpretation of simplified cEEG - total training duration 1day. The ICU physicians then interpreted 71 simplified cEEG recordings from 37 comatose survivors of cardiac arrest. The cEEG included amplitude-integrated EEG trends and two channels with original EEG-signals. Basic EEG background patterns and presence of epileptiform discharges or seizure activity were assessed on 5-grade rank-ordered scales based on standardized EEG terminology. An EEG-expert was used as reference. There was substantial agreement (κ 0.69) for EEG background patterns and moderate agreement (κ 0.43) for epileptiform discharges between ICU physicians and the EEG-expert. Sensitivity for detecting seizure activity by ICU physicians was limited (50%), but with high specificity (87%). After cardiac arrest, preliminary bedside interpretations of simplified cEEGs by trained ICU physicians may allow earlier detection of clinically relevant cEEG changes, prompting changes in patient management as well as additional evaluation by an EEG-expert. This strategy requires awareness of limitations of both the simplified electrode montage and the cEEG interpretations performed by ICU physicians. cEEG evaluation by an expert should not be delayed.

Taking the EEG Back Into the Brain: The Power of Multiple Discrete Sources.

Background: In contrast to many neuroimaging modalities, clinical interpretation of EEG does not take advantage of post-processing and digital signal analysis. In most centers, EEG is still interpreted at sensor level, exactly as half a century ago. A major task in clinical EEG interpretation is the identification of interictal epileptiform discharges (IEDs). However, due to the overlap of background activity, IEDs can be hard to detect in the scalp EEG. Since traditional montages, like bipolar and average reference, are linear transformations of the recorded channels, the question is whether we can provide linear transformations of the digital EEG to take it back into the brain, at least on a macroscopic level. The goal is to improve visibility of epileptiform activities and to separate out most of the overlap.Methods: Multiple discrete sources provide a stable linear inverse to transform the EEG into source space with little cross-talk between source regions. The model can be based on a few dipoles or regional sources, adapted to the individual EEG and MRI data, or on selected standard sources evenly distributed throughout the brain, e.g. below the 25 EEG standard electrodes.Results: Auditory and somatosensory evoked potentials serve as teaching examples to show how various source spaces can reveal the underlying source components including their loss or alteration due to lesions. Source spaces were able to reveal the propagation of source activities in frontal IEDs and the sequential activation of the major nodes of the underlying epileptic network in myoclonic epilepsy. The power of multiple discrete sources in separating the activities of different brain regions was also evident in the ongoing EEG of cases with frontal cortical dysplasia and bitemporal lobe epilepsy. The new source space 25 made IEDs more clearly visible over the EEG background signals. The more focal nature of source vs. scalp space was quantitatively confirmed using a new measurement of focality.Conclusion: Multiple discrete sources have the power to transform the EEG back into the brain by defining new EEG traces in source space. Using standard source space 25, these can provide for improved clinical interpretation of EEG.

Exploiting Graphoelements and Convolutional Neural Networks with Long Short Term Memory for Classification of the Human Electroencephalogram

The electroencephalogram (EEG) is a cornerstone of neurophysiological research and clinical neurology. Historically, the classification of EEG as showing normal physiological or abnormal pathological activity has been performed by expert visual review. The potential value of unbiased, automated EEG classification has long been recognized, and in recent years the application of machine learning methods has received significant attention. A variety of solutions using convolutional neural networks (CNN) for EEG classification have emerged with impressive results. However, interpretation of CNN results and their connection with underlying basic electrophysiology has been unclear. This paper proposes a CNN architecture, which enables interpretation of intracranial EEG (iEEG) transients driving classification of brain activity as normal, pathological or artifactual. The goal is accomplished using CNN with long short-term memory (LSTM). We show that the method allows the visualization of iEEG graphoelements with the highest contribution to the final classification result using a classification heatmap and thus enables review of the raw iEEG data and interpret the decision of the model by electrophysiology means.

LC7. Artifacts recognition and pitfalls in EEG rerecording and interpretation

LC7. Artifacts recognition and pitfalls in EEG rerecording and interpretation

P2-01-05. A trial of internal quality control of interpretation of EEG by establishing quantitative evaluation methods

P2-01-05. A trial of internal quality control of interpretation of EEG by establishing quantitative evaluation methods

Quality of EEG recordings in neurology clinics in Iran.

The aim of this study was to investigate the pitfalls in performing EEG in neurology centres in Iran, in order to identify any mistakes and provide appropriate advice for the neurology community. In this cross-sectional study, all patients who referred to the outpatient epilepsy clinic at Shiraz University of Medical Sciences, from April through to June 2018, were asked to provide any EEGs they had from other neurology centres. We reviewed the following information from their EEGs: number of channels, the design of the montages, and the setting of the filters. Eighty EEGs from different neurology centres were studied. With respect to the number of channels, 27 centres (34%) recorded 16 or more channels. With regards to the montage design, 34 centres (43%) had an appropriate design(s). With respect to high-frequency filter, three centres (4%) recorded their EEGs with 70-Hz high filter. With regards to the low-frequency filter, 43 centres (54%) recorded their EEGs with 1-Hz or lower filter. Just one centre performed EEGs in a standard fashion considering all variables. In this study, we observed that most neurology centres in Iran did not record EEGs in a standard and acceptable fashion. Provision of a well-designed educational programme that ensures neurology residents are competent in the fundamentals of EEG recording and interpretation is of paramount significance.

EEG-based outcome prediction after cardiac arrest with convolutional neural networks: Performance and visualization of discriminative features.

Prognostication for comatose patients after cardiac arrest is a difficult but essential task. Currently, visual interpretation of electroencephalogram (EEG) is one of the main modality used in outcome prediction. There is a growing interest in computer-assisted EEG interpretation, either to overcome the possible subjectivity of visual interpretation, or to identify complex features of the EEG signal. We used a one-dimensional convolutional neural network (CNN) to predict functional outcome based on 19-channel-EEG recorded from 267 adult comatose patients during targeted temperature management after CA. The area under the receiver operating characteristic curve (AUC) on the test set was 0.885. Interestingly, model architecture and fine-tuning only played a marginal role in classification performance. We then used gradient-weighted class activation mapping (Grad-CAM) as visualization technique to identify which EEG features were used by the network to classify an EEG epoch as favorable or unfavorable outcome, and also to understand failures of the network. Grad-CAM showed that the network relied on similar features than classical visual analysis for predicting unfavorable outcome (suppressed background, epileptiform transients). This study confirms that CNNs are promising models for EEG-based prognostication in comatose patients, and that Grad-CAM can provide explanation for the models' decision-making, which is of utmost importance for future use of deep learning models in a clinical setting.

Automatic detection and sonification of nonmotor generalized onset epileptic seizures: Preliminary results

Long-term video-EEG monitoring has improved diagnosis and treatment of epilepsy, especially in children. However, the amount of data neurophysiologists must analyze has grown remarkably.The main purpose of this paper is to provide a diagnostic support to speed up and ease EEG interpretation for a specific application concerning absence seizures, a type of non-motor generalized epileptic seizures.The proposed method consists of a pre-processing step where signals are filtered through the Stationary Wavelet Transform for the reduction of possible artefacts. Subsequently, a supervised automatic classification method is implemented for seizure detection, based on the Support Vector Machine Fine Gaussian method. Finally, a post-processing step is implemented in which spatial and temporal thresholds are defined for both online and offline application.In addition, a method that applies sonification techniques is developed. Sonification techniques could speed up the process of interpreting information, allowing rapid clinical intervention and a continuous monitoring of the event.The dataset consists of 30 EEG recordings performed in 24 children with absence seizures, clinically evaluated at the Meyer Children's Hospital in Firenze, Italy.The method shows encouraging results both in terms of balanced accuracy (about 96%) and latency times (1.25 s on average), which might make it suitable for online clinical trials. In fact, it was implemented in the perspective of a possible real-time application in clinical practice.

Normal Awake, Drowsy, and Sleep EEG Patterns That Might Be Overinterpreted as Abnormal.

Knowledge of normal patterns is essential for correct EEG interpretation. The overinterpretation of EEG (i.e., ascribing abnormality to EEG patterns that are not associated with disease) is a common problem and can contribute to misdiagnosis and mismanagement. Here, the authors concisely review normal patterns that might be improperly interpreted as abnormal. These include posterior slow waves of youth, central theta, K complexes, asymmetric sleep spindles, hypnagogic and hypnopompic hypersynchrony, arousal patterns, rhythmic midtemporal theta of drowsiness, and the wicket rhythm. Recognition of these patterns will lead to greater accuracy in EEG interpretation and help avoid incorrect management.

Initial experience of a telemetry EEG amplifier (Headset™) in the emergent diagnosis of nonconvulsive status epilepticus

Objectives: In the diagnosis of nonconvulsive status epilepticus (NCSE), the use of ongoing ictal electroencephalographic (EEG) findings is the gold standard. However, this is difficult without the availability of continuous EEG monitoring facilities. Given this, a small telemetry EEG amplifier (Headset-EEG; AE-120A EEG Headset™) was introduced for use in simple EEG monitoring facilities. In this study, we examined the diagnostic performance of the Headset-EEG for NCSE in a neuroemergency setting.Methods: We retrospectively selected five patients who were diagnosed with NCSE using Headset-EEG by attending physicians between October 2017 and July 2018. Reevaluation with Headset-EEG was performed, and the findings were compared with the initial diagnosis. Results: In Case 1, Headset-EEG clearly revealed repeated ictal discharges associated with acute subdural hematoma. However, in Cases 2 and 3, the initial EEG diagnosis was incorrectly made because of an EEG interpretation error. In Cases 4 and 5, various artifacts, caused by unstable electrode placement, resulted in an initial diagnosis of ongoing seizure activity. Nevertheless, the initial diagnoses and subsequent treatment of Cases 2–5 did not directly affect the patients' outcomes.Conclusion: The present study indicates that Headset-EEG is a possible tool for use in NCSE screening. However, considerable knowledge of EEG interpretation is needed to establish a correct NCSE diagnosis. When NCSE is suspected per Headset-EEG, a subsequent routine EEG with a full montage of 10–20 EEG electrodes should be performed in collaboration with an epileptologist or electroencephalographer.

Smartphone EEG and remote online interpretation for children with epilepsy in the Republic of Guinea: Quality, characteristics, and practice implications

Purpose: Children with epilepsy in low-income countries often go undiagnosed and untreated. We examine a portable, low-cost smartphone-based EEG technology in a heterogeneous pediatric epilepsy cohort in the West African Republic of Guinea. Methods: Children with epilepsy were recruited at the Ignace Deen Hospital in Conakry, 2017. Participants underwent sequential EEG recordings with an app-based EEG, the Smartphone Brain Scanner-2 (SBS2) and a standard Xltek EEG. Raw EEG data were transmitted via Bluetooth™ connection to an Android™ tablet and uploaded for remote EEG specialist review and reporting via a new, secure web-based reading platform, crowdEEG. The results were compared to same-visit Xltek 10–20 EEG recordings for identification of epileptiform and non-epileptiform abnormalities. Results: 97 children meeting the International League Against Epilepsy’s definition of epilepsy (49 male; mean age 10.3 years, 29 untreated with an antiepileptic drug; 0 with a prior EEG) were enrolled. Epileptiform discharges were detected on 21 (25.3%) SBS2 and 31 (37.3%) standard EEG recordings. The SBS2 had a sensitivity of 51.6% (95%CI 32.4%, 70.8%) and a specificity of 90.4% (95%CI 81.4%, 94.4%) for all types of epileptiform discharges, with positive and negative predictive values of 76.2% and 75.8% respectively. For generalized discharges, the SBS2 had a sensitivity of 43.5% with a specificity of 96.2%. Conclusions: The SBS2 has a moderate sensitivity and high specificity for the detection of epileptiform abnormalities in children with epilepsy in this low-income setting. Use of the SBS2+crowdEEG platform permits specialist input for patients with previously poor access to clinical neurophysiology expertise.