Reduced electrode arrays for the automated detection of rhythmic and periodic patterns in the intensive care unit: Frequently tried, frequently failed?

Automatic Detection and Classification of Epileptic Seizures from EEG Data: Finding Optimal Acquisition Settings and Testing Interpretable Machine Learning Approach

Deep learning (DL) is emerging as a successful technique for automatic detection and differentiation of spontaneous seizures that may otherwise be missed or misclassified. Herein, we propose a system architecture based on top-performing DL models for binary and multigroup classifications with the non-overlapping window technique, which we tested on the TUSZ dataset. The system accurately detects seizure episodes (87.7% Sn, 91.16% Sp) and carefully distinguishes eight seizure types (95-100% Acc). An increase in EEG sampling rate from 50 to 250 Hz boosted model performance: the precision of seizure detection rose by 5%, and seizure differentiation by 7%. A low sampling rate is a reasonable solution for training reliable models with EEG data. Decreasing the number of EEG electrodes from 21 to 8 did not affect seizure detection but worsened seizure differentiation significantly: 98.24 ± 0.17 vs. 85.14 ± 3.14% recall. In detecting epileptic episodes, all electrodes provided equally informative input, but in seizure differentiation, their informative value varied. We improved model explainability with interpretable ML. Activation maximization highlighted the presence of EEG patterns specific to eight seizure types. Cortical projection of epileptic sources depicted differences between generalized and focal seizures. Interpretable ML techniques confirmed that our system recognizes biologically meaningful features as indicators of epileptic activity in EEG.

Seizure detection with reduced electroencephalogram channels: research trends and outlook

Epilepsy is a prevalent condition characterized by recurrent, unpredictable seizures. Monitoring with surface electroencephalography (EEG) is the gold standard for diagnosing epilepsy, but a time-consuming, uncomfortable and sometimes ineffective process for patients. Further, using EEG over a brief monitoring period has variable success, dependent on patient tolerance and seizure frequency. The availability of hospital resources and hardware and software specifications inherently restrict the options for comfortable, long-term data collection, resulting in limited data for training machine-learning models. This mini-review examines the current patient journey, providing an overview of the current state of EEG monitoring with reduced electrodes and automated channel reduction methods. Opportunities for improving data reliability through multi-modal data fusion are suggested. We assert the need for further research in electrode reduction to advance brain monitoring solutions towards portable, reliable devices that simultaneously offer patient comfort, perform ultra-long-term monitoring and expedite the diagnosis process.

The Impact of Nonconvulsive Status Epilepticus after Cardiac Surgery on Outcome

Neurological complications after heart surgery are associated with tremendous morbidity and mortality. Nonconvulsive status epilepticus (NCSE), which can only be verified by EEG, may cause secondary brain damage. Its frequency and its impact on outcomes after cardiac surgery is still unclear. We collected the neurological files and clinical data of all our patients after heart surgery who, in the course of their ICU stay, had been seen by a neurologist who ordered an EEG. Within 18 months, 1457 patients had cardiac surgery on cardiopulmonary bypass. EEG was requested for 89 patients. Seizures were detected in 39 patients and NCSE was detected in 11 patients. Open heart surgery was performed in all 11 NSCE patients, of whom eight showed concomitant brain insults. None had a history of epilepsy. Despite the inhibition of seizure activity with antiseizure medication, clinical improvement was only noted in seven NCSE patients, three of whom were in cerebral performance category 2 and four in category 3 at hospital discharge. The four patients without neurological benefit subsequently died in the ICU. The occurrence of NCSE after open cardiac surgery is significant and frequently associated with brain injury. It seems prudent to perform EEG studies early to interrupt seizure activity and mitigate secondary cerebral injury.

A Comparison of Energy-Efficient Seizure Detectors for Implantable Neurostimulation Devices

Introduction

About 30% of epilepsy patients are resistant to treatment with antiepileptic drugs, and only a minority of these are surgical candidates. A recent therapeutic approach is the application of electrical stimulation in the early phases of a seizure to interrupt its spread across the brain. To accomplish this, energy-efficient seizure detectors are required that are able to detect a seizure in its early stages.

Methods

Three patient-specific, energy-efficient seizure detectors are proposed in this study: (i) random forest (RF); (ii) long short-term memory (LSTM) recurrent neural network (RNN); and (iii) convolutional neural network (CNN). Performance evaluation was based on EEG data (n = 40 patients) derived from a selected set of surface EEG electrodes, which mimic the electrode layout of an implantable neurostimulation system. As for the RF input, 16 features in the time- and frequency-domains were selected. Raw EEG data were used for both CNN and RNN. Energy consumption was estimated by a platform-independent model based on the number of arithmetic operations (AOs) and memory accesses (MAs). To validate the estimated energy consumption, the RNN classifier was implemented on an ultra-low-power microcontroller.

Results

The RNN seizure detector achieved a slightly better level of performance, with a median area under the precision-recall curve score of 0.49, compared to 0.47 for CNN and 0.46 for RF. In terms of energy consumption, RF was the most efficient algorithm, with a total of 67k AOs and 67k MAs per classification. This was followed by CNN (488k AOs and 963k MAs) and RNN (772k AOs and 978k MAs), whereby MAs contributed more to total energy consumption. Measurements derived from the hardware implementation of the RNN algorithm demonstrated a significant correlation between estimations and actual measurements.

Discussion

All three proposed seizure detection algorithms were shown to be suitable for application in implantable devices. The applied methodology for a platform-independent energy estimation was proven to be accurate by way of hardware implementation of the RNN algorithm. These findings show that seizure detection can be achieved using just a few channels with limited spatial distribution. The methodology proposed in this study can therefore be applied when designing new models for responsive neurostimulation.

Limited-Montage EEG as a Tool for the Detection of Nonconvulsive Seizures

PURPOSE

Prefabricated arrays with a limited number of electrodes offer an opportunity to hasten the diagnosis of seizures; however, their accuracy to detect seizures is unknown. We examined the utility of two limited-montage EEG setups for the detection of nonconvulsive seizures.

METHODS

Thirty previously interpreted EEG segments with nonconvulsive seizures from 30 patients and 60 segments with background slowing or normal EEG from 60 patients were rendered in a bipolar "double banana" montage, a double distance "neonatal" montage, and a circumferential "hatband" montage. Experts reviewed 60 to 180 seconds long segments to determine whether seizures were present and if the EEG data provided were sufficient to make a decision on escalation of clinical care by ordering an additional EEG or prescribing anticonvulsants. The periodic patterns on the ictal-interictal continuum were specifically excluded for this analysis to keep the focus on definite electrographic seizures.

RESULTS

The sensitivities for seizure of the neonatal and hatband montages were 0.96 and 0.84, respectively, when compared with full montage EEG, whereas the specificities were 0.94 and 0.98, respectively. Appropriate escalation of care was suggested for 96% and 92% of occurrences of seizure patterns in neonatal and hatband montages, respectively. When compared with clinical EEG, the sensitivities of the neonatal and hatband montages for seizure diagnosis were 0.85 and 0.69, respectively.

CONCLUSIONS

Nonconvulsive seizures were detected with high accuracy using the limited electrode array configuration in the neonatal and hatband montages. The sensitivity of the neonatal montage EEG in detecting seizures was superior to that of a hatband montage. These findings suggest that in some patients with nonconvulsive seizures, limited-montage EEG may allow to differentiate ictal and slow patterns.

Continuous vs Routine Electroencephalogram in Critically Ill Adults With Altered Consciousness and No Recent Seizure: A Multicenter Randomized Clinical Trial

Question

In patients with acute consciousness impairment and no recent seizures, does continuous electroencephalogram (cEEG) correlate with reduced mortality compared with repeated routine EEG (rEEG)?

Findings

In this pragmatic, multicenter randomized clinical trial analyzing 364 adults, cEEG translated into a higher rate of seizures/status epilepticus detection and antiseizure treatment modifications but did not improve mortality compared with rEEG.

Meaning

Pending larger studies, rEEG may represent a valid alternative to cEEG in centers with limited resources.

Importance

In critically ill patients with altered consciousness, continuous electroencephalogram (cEEG) improves seizure detection, but is resource-consuming compared with routine EEG (rEEG). It is also uncertain whether cEEG has an effect on outcome.

Objective

To assess whether cEEG is associated with reduced mortality compared with rEEG.

Design, Setting, and Participants

The pragmatic multicenter Continuous EEG Randomized Trial in Adults (CERTA) was conducted between 2017 and 2018, with follow-up of 6 months. Outcomes were assessed by interviewers blinded to interventions.The study took place at 4 tertiary hospitals in Switzerland (intensive and intermediate care units). Depending on investigators’ availability, we pragmatically recruited critically ill adults having Glasgow Coma Scale scores of 11 or less or Full Outline of Responsiveness score of 12 or less, without recent seizures or status epilepticus. They had cerebral (eg, brain trauma, cardiac arrest, hemorrhage, or stroke) or noncerebral conditions (eg, toxic-metabolic or unknown etiology), and EEG was requested as part of standard care. An independent physician provided emergency informed consent.

Interventions

Participants were randomized 1:1 to cEEG for 30 to 48 hours vs 2 rEEGs (20 minutes each), interpreted according to standardized American Clinical Neurophysiology Society guidelines.

Main Outcomes and Measures

Mortality at 6 months represented the primary outcome. Secondary outcomes included interictal and ictal features detection and change in therapy.

Results

We analyzed 364 patients (33% women; mean [SD] age, 63 [15] years). At 6 months, mortality was 89 of 182 in those with cEEG and 88 of 182 in those with rEEG (adjusted relative risk [RR], 1.02; 95% CI, 0.83-1.26; P = .85). Exploratory comparisons within subgroups stratifying patients according to age, premorbid disability, comorbidities on admission, deeper consciousness reduction, and underlying diagnoses revealed no significant effect modification. Continuous EEG was associated with increased detection of interictal features and seizures (adjusted RR, 1.26; 95% CI, 1.08-1.15; P = .004 and 3.37; 95% CI, 1.63-7.00; P = .001, respectively) and more frequent adaptations in antiseizure therapy (RR, 1.84; 95% CI, 1.12-3.00; P = .01).

Conclusions and Relevance

This pragmatic trial shows that in critically ill adults with impaired consciousness and no recent seizure, cEEG leads to increased seizure detection and modification of antiseizure treatment but is not related to improved outcome compared with repeated rEEG. Pending larger studies, rEEG may represent a valid alternative to cEEG in centers with limited resources.

Trial Registration

ClinicalTrials.gov Identifier: NCT03129438

Virtual EEG-electrodes: Convolutional neural networks as a method for upsampling or restoring channels

BACKGROUND

In clinical practice, EEGs are assessed visually. For practical reasons, recordings often need to be performed with a reduced number of electrodes and artifacts make assessment difficult. To circumvent these obstacles, different interpolation techniques can be utilized. These techniques usually perform better for higher electrode densities and values interpolated at areas far from electrodes can be unreliable. Using a method that learns the statistical distribution of the cortical electrical fields and predicts values may yield better results.

NEW METHOD

Generative networks based on convolutional layers were trained to upsample from 4 or 14 channels or to dynamically restore single missing channels to recreate 21-channel EEGs. 5,144 h of data from 1,385 subjects of the Temple University Hospital EEG database were used for training and evaluating the networks.

COMPARISON WITH EXISTING METHOD

The results were compared to spherical spline interpolation. Several statistical measures were used as well as a visual evaluation by board certified clinical neurophysiologists. Overall, the generative networks performed significantly better. There was no difference between real and network generated data in the number of examples assessed as artificial by experienced EEG interpreters whereas for data generated by interpolation, the number was significantly higher. In addition, network performance improved with increasing number of included subjects, with the greatest effect seen in the range 5-100 subjects.

CONCLUSIONS

Using neural networks to restore or upsample EEG signals is a viable alternative to spherical spline interpolation.

Standardized visual EEG features predict outcome in patients with acute consciousness impairment of various etiologies

BACKGROUND

Early prognostication in patients with acute consciousness impairment is a challenging but essential task. Current prognostic guidelines vary with the underlying etiology. In particular, electroencephalography (EEG) is the most important paraclinical examination tool in patients with hypoxic ischemic encephalopathy (HIE), whereas it is not routinely used for outcome prediction in patients with traumatic brain injury (TBI).

METHOD

Data from 364 critically ill patients with acute consciousness impairment (GCS ≤ 11 or FOUR ≤ 12) of various etiologies and without recent signs of seizures from a prospective randomized trial were retrospectively analyzed. Random forest classifiers were trained using 8 visual EEG features-first alone, then in combination with clinical features-to predict survival at 6 months or favorable functional outcome (defined as cerebral performance category 1-2).

RESULTS

The area under the ROC curve was 0.812 for predicting survival and 0.790 for predicting favorable outcome using EEG features. Adding clinical features did not improve the overall performance of the classifier (for survival: AUC = 0.806, p = 0.926; for favorable outcome: AUC = 0.777, p = 0.844). Survival could be predicted in all etiology groups: the AUC was 0.958 for patients with HIE, 0.955 for patients with TBI and other neurosurgical diagnoses, 0.697 for patients with metabolic, inflammatory or infectious causes for consciousness impairment and 0.695 for patients with stroke. Training the classifier separately on subgroups of patients with a given etiology (and thus using less training data) leads to poorer classification performance.

CONCLUSIONS

While prognostication was best for patients with HIE and TBI, our study demonstrates that similar EEG criteria can be used in patients with various causes of consciousness impairment, and that the size of the training set is more important than homogeneity of ACI etiology.

Machine Learning Applications in the Neuro ICU: A Solution to Big Data Mayhem?

The neurological ICU (neuro ICU) often suffers from significant limitations due to scarce resource availability for their neurocritical care patients. Neuro ICU patients require frequent neurological evaluations, continuous monitoring of various physiological parameters, frequent imaging, and routine lab testing. This amasses large amounts of data specific to each patient. Neuro ICU teams are often overburdened by the resulting complexity of data for each patient. Machine Learning algorithms (ML), are uniquely capable of interpreting high-dimensional datasets that are too difficult for humans to comprehend. Therefore, the application of ML in the neuro ICU could alleviate the burden of analyzing big datasets for each patient. This review serves to (1) briefly summarize ML and compare the different types of MLs, (2) review recent ML applications to improve neuro ICU management and (3) describe the future implications of ML to neuro ICU management.

Diagnostic Value of Electroencephalography with Ten Electrodes in Critically Ill Patients

BACKGROUND

In critical care settings, electroencephalography (EEG) with reduced number of electrodes (reduced montage EEG, rm-EEG) might be a timely alternative to the conventional full montage EEG (fm-EEG). However, past studies have reported variable accuracies for detecting seizures using rm-EEG. We hypothesized that the past studies did not distinguish between differences in sensitivity from differences in classification of EEG patterns by different readers. The goal of the present study was to revisit the diagnostic value of rm-EEG when confounding issues are accounted for.

METHODS

We retrospectively collected 212 adult EEGs recorded at Massachusetts General Hospital and reviewed by two epileptologists with access to clinical, trending, and video information. In Phase I of the study, we re-configured the first 4 h of the EEGs in lateral circumferential montage with ten electrodes and asked new readers to interpret the EEGs without access to any other ancillary information. We compared their rating to the reading of hospital clinicians with access to ancillary information. In Phase II, we measured the accuracy of the same raters reading representative samples of the discordant EEGs in full and reduced configurations presented randomly by comparing their performance to majority consensus as the gold standard.

RESULTS

Of the 95 EEGs without seizures in the selected fm-EEG, readers of rm-EEG identified 92 cases (97%) as having no seizure activity. Of 117 EEGs with "seizures" identified in the selected fm-EEG, none of the cases was labeled as normal on rm-EEG. Readers of rm-EEG reported pathological activity in 100% of cases, but labeled them as seizures (N = 77), rhythmic or periodic patterns (N = 24), epileptiform spikes (N = 7), or burst suppression (N = 6). When the same raters read representative epochs of the discordant EEG cases (N = 43) in both fm-EEG and rm-EEG configurations, we found high concordance (95%) and intra-rater agreement (93%) between fm-EEG and rm-EEG diagnoses.

CONCLUSIONS

Reduced EEG with ten electrodes in circumferential configuration preserves key features of the traditional EEG system. Discrepancies between rm-EEG and fm-EEG as reported in some of the past studies can be in part due to methodological factors such as choice of gold standard diagnosis, asymmetric access to ancillary clinical information, and inter-rater variability rather than detection failure of rm-EEG as a result of electrode reduction per se.

Comparison of Frontal-Temporal Channels in Epilepsy Seizure Prediction Based on EEMD-ReliefF and DNN

Epilepsy patients who do not have their seizures controlled with medication or surgery live in constant fear. The psychological burden of uncertainty surrounding the occurrence of random seizures is one of the most stressful and debilitating aspects of the disease. Despite the research progress in this field, there is a need for a non-invasive prediction system that helps disrupt the seizure epileptiform. Electroencephalogram (EEG) signals are non-stationary, nonlinear and vary with each patient and every recording. Full use of the non-invasive electrode channels is impractical for real-time use. We propose two frontal-temporal electrode channels based on ensemble empirical mode decomposition (EEMD) and Relief methods to address these challenges. The EEMD decomposes the segmented data frame in the ictal state into its intrinsic mode functions, and then we apply Relief to select the most relevant oscillatory components. A deep neural network (DNN) model learns these features to perform seizure prediction and early detection of patient-specific EEG recordings. The model yields an average sensitivity and specificity of 86.7% and 89.5%, respectively. The two-channel model shows the ability to capture patterns from brain locations for non-fontal-temporal seizures.

From research to clinic: A sensor reduction method for high-density EEG neurofeedback systems

OBJECTIVE

To accurately deliver a source-estimated neurofeedback (NF) signal developed on a 128-sensors EEG system on a reduced 32-sensors EEG system.

METHODS

A linearly constrained minimum variance beamformer algorithm was used to select the 64 sensors which contributed most highly to the source signal. Monte Carlo-based sampling was then used to randomly generate a large set of reduced 32-sensors montages from the 64 beamformer-selected sensors. The reduced montages were then tested for their ability to reproduce the 128-sensors NF. The high-performing montages were then pooled and analyzed by a k-means clustering machine learning algorithm to produce an optimized reduced 32-sensors montage.

RESULTS

Nearly 4500 high-performing montages were discovered from the Monte Carlo sampling. After statistically analyzing this pool of high performing montages, a set of refined 32-sensors montages was generated that could reproduce the 128-sensors NF with greater than 80% accuracy for 72% of the test population.

CONCLUSION

Our Monte Carlo reduction method was used to create reliable reduced-sensors montages which could be used to deliver accurate NF in clinical settings.

SIGNIFICANCE

A translational pathway is now available by which high-density EEG-based NF measures can be delivered using clinically accessible low-density EEG systems.

Sensitivity of a Reduced EEG Montage for Seizure Detection in the Neurocritical Care Setting

INTRODUCTION

Neurocritical care units commonly implement the double-distance reduced EEG montage in postoperative neurosurgic patients who have structural barriers that hinder the placement of a standard 10-20 system array. Despite its widespread use, its sensitivity has not been adequately addressed. We evaluated the sensitivity and specificity of this montage for seizure detection.

METHODS

One hundred fifty-five full-montage continuous EEGs (cEEGs) completed in the Johns Hopkins University neurocritical care unit containing unequivocal electrographic seizures, status epilepticus, or other abnormalities were selected, comprising 73 ictal and 82 nonictal EEGs. EEGs were reformatted to the reduced montage, and 2-hour clips were reviewed independently by 2 epileptologists who documented the presence of seizures, status, or background abnormalities.

RESULTS

The sensitivity and specificity of the reduced montage for electrographic seizure detection was 81% and 92% with substantial interrater agreement (kappa 0.71). The sensitivity for status epilepticus was lower at 69%, but specificity remained high at 97% (kappa 0.67). Several EEGs miscategorized as nonictal were labeled as rather having rhythmic activity or periodic discharges. Evaluation of background patterns on the ictal-interictal continuum resulted in sensitivities ranging from 68% to 83%.

CONCLUSIONS

Although the specificity of the reduced array is good, epileptologists should remain vigilant when monitoring patients using this montage, given its reduced sensitivity for epileptic activity, especially status epilepticus.

Automatic Computer-Based Detection of Epileptic Seizures

Automatic computer-based seizure detection and warning devices are important for objective seizure documentation, for SUDEP prevention, to avoid seizure related injuries and social embarrassments as a consequence of seizures, and to develop on demand epilepsy therapies. Automatic seizure detection systems can be based on direct analysis of epileptiform discharges on scalp-EEG or intracranial EEG, on the detection of motor manifestations of epileptic seizures using surface electromyography (sEMG), accelerometry (ACM), video detection systems and mattress sensors and finally on the assessment of changes of physiologic parameters accompanying epileptic seizures measured by electrocardiography (ECG), respiratory monitors, pulse oximetry, surface temperature sensors, and electrodermal activity. Here we review automatic seizure detection based on scalp-EEG, ECG, and sEMG. Different seizure types affect preferentially different measurement parameters. While EEG changes accompany all types of seizures, sEMG and ACM are suitable mainly for detection of seizures with major motor manifestations. Therefore, seizure detection can be optimized by multimodal systems combining several measurement parameters. While most systems provide sensitivities over 70%, specificity expressed as false alarm rates still needs to be improved. Patients' acceptance and comfort of a specific device are of critical importance for its long-term application in a meaningful clinical way.

Diagnostic utility of eight-channel EEG for detecting generalized or hemispheric seizures and rhythmic periodic patterns

•

Current practice lacks rapid detection tools to screen for seizures.

•

High agreement exists between neurologists’ diagnoses using full and reduced montage EEG.

•

Reduced channel EEG can be used to screen for generalized or hemispheric or rhythmic and periodic abnormalities.

Objectives

To compare the diagnostic utility of electroencephalography (EEG) using reduced, 8-channel montage (rm-EEG) to full, 18-channel montage (fm-EEG) for detection of generalized or hemispheric seizures and rhythmic periodic patterns (RPPs) by neurologists with extensive EEG training, neurology residents with minimal EEG exposure, and medical students without EEG experience.

Methods

We presented EEG samples in both fm-EEG (bipolar montage) and rm-EEG (lateral leads of bipolar montage) to 20 neurologists, 20 residents, and 42 medical students. Unanimous agreement of three senior epileptologists defined samples as seizures (n = 7), RPPs (n = 10), and normal or slowing (n = 20). Differences in median accuracy, sensitivity, and specificity were assessed using Wilcoxon signed-rank tests.

Results

Full and reduced EEG demonstrated similar accuracy when read by neurologists (fm-EEG: 95%, rm-EEG: 95%, p = 0.29), residents (fm-EEG: 80%, rm-EEG: 80%, p = 0.05), and students (fm-EEG: 60%, rm-EEG: 51%, p = 0.68). Moreover, neurologists’ sensitivity for detecting seizure activity was comparable between fm-EEG (100%) and rm-EEG (98%) (p = 0.17). Furthermore, the specificity of rm-EEG for seizures and RPP (neurologists: 100%, residents: 90%, students: 86%) was significantly greater than that of fm-EEG (neurologists: 93%, p = 0.03; residents: 80%, p = 0.01; students: 69%, p < 0.001).

Conclusions

The reduction of the number of EEG channels from 18 to 8 does not compromise neurologists’ sensitivity for detecting seizures that are often a core reason for performing urgent EEG. It may also increase their specificity for detecting rhythmic and periodic patterns, and thereby providing important diagnostic information to guide patient’s management.

Significance

Our study is the first to document the utility of a reduced channel EEG above the hairline compared to full montage EEG in aiding medical staff with varying degrees of EEG training to detect generalized or hemispheric seizures.

Automatic multimodal detection for long-term seizure documentation in epilepsy

OBJECTIVE

This study investigated sensitivity and false detection rate of a multimodal automatic seizure detection algorithm and the applicability to reduced electrode montages for long-term seizure documentation in epilepsy patients.

METHODS

An automatic seizure detection algorithm based on EEG, EMG, and ECG signals was developed. EEG/ECG recordings of 92 patients from two epilepsy monitoring units including 494 seizures were used to assess detection performance. EMG data were extracted by bandpass filtering of EEG signals. Sensitivity and false detection rate were evaluated for each signal modality and for reduced electrode montages.

RESULTS

All focal seizures evolving to bilateral tonic-clonic (BTCS, n=50) and 89% of focal seizures (FS, n=139) were detected. Average sensitivity in temporal lobe epilepsy (TLE) patients was 94% and 74% in extratemporal lobe epilepsy (XTLE) patients. Overall detection sensitivity was 86%. Average false detection rate was 12.8 false detections in 24h (FD/24h) for TLE and 22 FD/24h in XTLE patients. Utilization of 8 frontal and temporal electrodes reduced average sensitivity from 86% to 81%.

CONCLUSION

Our automatic multimodal seizure detection algorithm shows high sensitivity with full and reduced electrode montages.

SIGNIFICANCE

Evaluation of different signal modalities and electrode montages paces the way for semi-automatic seizure documentation systems.