Estimating the likelihood of epilepsy from clinically noncontributory electroencephalograms using computational analysis: A retrospective, multisite case-control study.

OBJECTIVE: This study was undertaken to validate a set of candidate biomarkers of seizure susceptibility in a retrospective, multisite case-control study, and to determine the robustness of these biomarkers derived from routinely collected electroencephalography (EEG) within a large cohort (both epilepsy and common alternative conditions such as nonepileptic attack disorder). METHODS: The database consisted of 814 EEG recordings from 648 subjects, collected from eight National Health Service sites across the UK. Clinically noncontributory EEG recordings were identified by an experienced clinical scientist (N = 281; 152 alternative conditions, 129 epilepsy). Eight computational markers (spectral [n = 2], network-based [n = 4], and model-based [n = 2]) were calculated within each recording. Ensemble-based classifiers were developed using a two-tier cross-validation approach. We used standard regression methods to assess whether potential confounding variables (e.g., age, gender, treatment status, comorbidity) impacted model performance. RESULTS: We found levels of balanced accuracy of 68% across the cohort with clinically noncontributory normal EEGs (sensitivity =61%, specificity =75%, positive predictive value =55%, negative predictive value =79%, diagnostic odds ratio =4.64, area under receiver operated characteristics curve =.72). Group level analysis found no evidence suggesting any of the potential confounding variables significantly impacted the overall performance. SIGNIFICANCE: These results provide evidence that the set of biomarkers could provide additional value to clinical decision-making, providing the foundation for a decision support tool that could reduce diagnostic delay and misdiagnosis rates. Future work should therefore assess the change in diagnostic yield and time to diagnosis when utilizing these biomarkers in carefully designed prospective studies.

Changes in the heartbeat-evoked potential are associated with functional seizures.

BACKGROUND: Patients with functional seizures (FS) can experience dissociation (depersonalisation) before their seizures. Depersonalisation reflects disembodiment, which may be related to changes in interoceptive processing. The heartbeat-evoked potential (HEP) is an electroencephalogram (EEG) marker of interoceptive processing. AIM: To assess whether alterations in interoceptive processing indexed by HEP occur prior to FS and compare this with epileptic seizures (ES). METHODS: HEP amplitudes were calculated from EEG during video-EEG monitoring in 25 patients with FS and 19 patients with ES, and were compared between interictal and preictal states. HEP amplitude difference was calculated as preictal HEP amplitude minus interictal HEP amplitude. A receiver operating characteristic (ROC) curve analysis was used to evaluate the diagnostic performance of HEP amplitude difference in discriminating FS from ES. RESULTS: The FS group demonstrated a significant reduction in HEP amplitude between interictal and preictal states at F8 (effect size rB=0.612, false discovery rate (FDR)-corrected q=0.030) and C4 (rB=0.600, FDR-corrected q=0.035). No differences in HEP amplitude were found between states in the ES group. Between diagnostic groups, HEP amplitude difference differed between the FS and ES groups at F8 (rB=0.423, FDR-corrected q=0.085) and C4 (rB=0.457, FDR-corrected q=0.085). Using HEP amplitude difference at frontal and central electrodes plus sex, we found that the ROC curve demonstrated an area under the curve of 0.893, with sensitivity=0.840 and specificity=0.842. CONCLUSION: Our data support the notion that aberrant interoception occurs prior to FS. Changes in HEP amplitude may reflect a neurophysiological biomarker of FS and may have diagnostic utility in differentiating FS and ES.

Lateralized Periodic Discharges During Remifentanil Infusion.

Lateralized periodic discharges (LPDs) are a common electroencephalographic (EEG) pattern in the neurointensive care unit setting. LPDs are typically observed in association with acute structural lesions of the brain with different etiologies. There are no reports describing a link between the occurrence of LPDs and the administration of remifentanil. Remifentanil is a rapid-acting pure µ-opioid receptor agonist, which is indicated to provide analgesia and sedation in mechanically ventilated patients in intensive care units. We present a case of an 84-year-old man with neuroglycopenia who developed LPDs while sedated with remifentanil. We report, for the first time, a potential relationship between remifentanil and the induction of LPDs.

Higher-order tensor decomposition based scalp-to-intracranial EEG projection for detection of interictal epileptiform discharges.

Objective.Interictal epileptiform discharges (IEDs) occur between two seizures onsets. IEDs are mainly captured by intracranial recordings and are often invisible over the scalp. This study proposes a model based on tensor factorization to map the time-frequency (TF) features of scalp EEG (sEEG) to the TF features of intracranial EEG (iEEG) in order to detect IEDs from over the scalp with high sensitivity.Approach.Continuous wavelet transform is employed to extract the TF features. Time, frequency, and channel modes of IED segments from iEEG recordings are concatenated into a four-way tensor. Tucker and CANDECOMP/PARAFAC decomposition techniques are employed to decompose the tensor into temporal, spectral, spatial, and segmental factors. Finally, TF features of both IED and non-IED segments from scalp recordings are projected onto the temporal components for classification.Main results.The model performance is obtained in two different approaches: within- and between-subject classification approaches. Our proposed method is compared with four other methods, namely a tensor-based spatial component analysis method, TF-based method, linear regression mapping model, and asymmetric-symmetric autoencoder mapping model followed by convolutional neural networks. Our proposed method outperforms all these methods in both within- and between-subject classification approaches by respectively achieving 84.2% and 72.6% accuracy values.Significance.The findings show that mapping sEEG to iEEG improves the performance of the scalp-based IED detection model. Furthermore, the tensor-based mapping model outperforms the autoencoder- and regression-based mapping models.

Guillain-Barré syndrome following COVID-19: a newly emerging post-infectious complication.

A 57-year-old man presented with a progressive flaccid symmetrical motor and sensory neuropathy following a 1-week history of cough and malaise. He was diagnosed with Guillain-Barré syndrome secondary to COVID-19 and started on intravenous immunoglobulin. He proceeded to have worsening respiratory function and needed intubation and mechanical ventilation. This is the first reported case of this rare neurological complication of COVID-19 in the UK, but it adds to a small but growing body of international evidence to suggest a significant association between these two conditions. Increasing appreciation of this by clinicians will ensure earlier diagnosis, monitoring and treatment of patients presenting with this.

The Single Word Auditory Comprehension (SWAC) test: A simple method to identify receptive language areas with electrical stimulation.

OBJECTIVES: Resective surgery for medically refractory epilepsy in proximity to speech receptive areas requires balancing adequate resection of the epileptogenic zone for optimal seizure control with preservation of function. We develop a simple test (Single Word Auditory Comprehension or SWAC) to localize speech receptive areas by evaluating patients' ability to comprehend a single word. METHODS: Patients were studied during presurgical or intraoperative assessment for epilepsy with intracranial electrodes. They were asked to listen to a common word (target word) and to describe what it meant without saying the target word. Electrical stimulation (trains of biphasic 2-ms pulses, 50 Hz for 3 s) was delivered while the patient listened to the target word, not while the patient explained the meaning of the word. In six patients, SWAC test was carried out during extraoperative chronic recordings, and in one patient in the operating theater under local anesthesia. RESULTS: Among the 7 patients where the test identified deficits, 6 underwent resection (4 temporal, 1 supramarginal, and 1 occipital). Two patients showed temporary minor speech deficits after resection. No patient showed permanent speech deficits after resection. CONCLUSION/SIGNIFICANCE: The SWAC test is reliable, simple and fast to implement, and suitable for intraoperating mapping. It could be used as a simple initial test to identify receptive language areas where more complex additional tests can be performed.

Deep Neural Architectures for Mapping Scalp to Intracranial EEG.

Data is often plagued by noise which encumbers machine learning of clinically useful biomarkers and electroencephalogram (EEG) data is no exemption. Intracranial EEG (iEEG) data enhances the training of deep learning models of the human brain, yet is often prohibitive due to the invasive recording process. A more convenient alternative is to record brain activity using scalp electrodes. However, the inherent noise associated with scalp EEG data often impedes the learning process of neural models, achieving substandard performance. Here, an ensemble deep learning architecture for nonlinearly mapping scalp to iEEG data is proposed. The proposed architecture exploits the information from a limited number of joint scalp-intracranial recording to establish a novel methodology for detecting the epileptic discharges from the sEEG of a general population of subjects. Statistical tests and qualitative analysis have revealed that the generated pseudo-intracranial data are highly correlated with the true intracranial data. This facilitated the detection of IEDs from the scalp recordings where such waveforms are not often visible. As a real-world clinical application, these pseudo-iEEGs are then used by a convolutional neural network for the automated classification of intracranial epileptic discharges (IEDs) and non-IED of trials in the context of epilepsy analysis. Although the aim of this work was to circumvent the unavailability of iEEG and the limitations of sEEG, we have achieved a classification accuracy of 68% an increase of 6% over the previously proposed linear regression mapping.

Characterizing EEG Cortical Dynamics and Connectivity with Responses to Single Pulse Electrical Stimulation (SPES).

OBJECTIVES: To model cortical connections in order to characterize their oscillatory behavior and role in the generation of spontaneous electroencephalogram (EEG). METHODS: We studied averaged responses to single pulse electrical stimulation (SPES) from the non-epileptogenic hemisphere of five patients assessed with intracranial EEG who became seizure free after contralateral temporal lobectomy. Second-order control system equations were modified to characterize the systems generating a given response. SPES responses were modeled as responses to a unit step input. EEG power spectrum was calculated on the 20[Formula: see text]s preceding SPES. RESULTS: 121 channels showed responses to 32 stimulation sites. A single system could model the response in 41.3% and two systems were required in 58.7%. Peaks in the frequency response of the models tended to occur within the frequency range of most activity on the spontaneous EEG. Discrepancies were noted between activity predicted by models and activity recorded in the spontaneous EEG. These discrepancies could be explained by the existence of alpha rhythm or interictal epileptiform discharges. CONCLUSIONS: Cortical interactions shown by SPES can be described as control systems which can predict cortical oscillatory behavior. The method is unique as it describes connectivity as well as dynamic interactions.

Detection of Interictal Discharges With Convolutional Neural Networks Using Discrete Ordered Multichannel Intracranial EEG.

Detection algorithms for electroencephalography (EEG) data, especially in the field of interictal epileptiform discharge (IED) detection, have traditionally employed handcrafted features, which utilized specific characteristics of neural responses. Although these algorithms achieve high accuracy, mere detection of an IED holds little clinical significance. In this paper, we consider deep learning for epileptic subjects to accommodate automatic feature generation from intracranial EEG data, while also providing clinical insight. Convolutional neural networks are trained in a subject independent fashion to demonstrate how meaningful features are automatically learned in a hierarchical process. We illustrate how the convolved filters in the deepest layers provide insight toward the different types of IEDs within the group, as confirmed by our expert clinicians. The morphology of the IEDs found in filters can help evaluate the treatment of a patient. To improve the learning of the deep model, moderately different score classes are utilized as opposed to binary IED and non-IED labels. The resulting model achieves state-of-the-art classification performance and is also invariant to time differences between the IEDs. This paper suggests that deep learning is suitable for automatic feature generation from intracranial EEG data, while also providing insight into the data.

The Role of Thalamus Versus Cortex in Epilepsy: Evidence from Human Ictal Centromedian Recordings in Patients Assessed for Deep Brain Stimulation.

BACKGROUND: The onset of generalized seizures is a long debated subject in epilepsy. The relative roles of cortex and thalamus in initiating and maintaining the different seizure types are unclear. OBJECTIVE: The purpose of the study is to estimate whether the cortex or the centromedian thalamic nucleus is leading in initiating and maintaining seizures in humans. METHODS: We report human ictal recordings with simultaneous thalamic and cortical electrodes from three patients without anesthesia being assessed for deep brain stimulation (DBS). Patients 1 and 2 had idiopathic generalized epilepsy whereas patient 3 had frontal lobe epilepsy. Visual inspection was combined with nonlinear correlation analysis. RESULTS: In patient 1, seizure onset was bilateral cortical and the belated onset of leading thalamic discharges was associated with an increase in rhythmicity of discharges, both in thalamus and cortex. In patient 2, we observed bilateral independent interictal discharges restricted to the thalamus. However, ictal onset was diffuse, with discharges larger in the cortex even though they were led by the thalamus. In patient 3, seizure onset was largely restricted to frontal structures, with belated lagging thalamic involvement. CONCLUSION: In human generalized seizures, the thalamus may become involved early or late in the seizure but, once it becomes involved, it leads the cortex. In contrast, in human frontal seizures the thalamus gets involved late in the seizure and, once it becomes involved, it lags behind the cortex. In addition, the centromedian nucleus of the thalamus is capable of autonomous epileptogenesis as suggested by the presence of independent focal unilateral epileptiform discharges restricted to thalamic structures. The thalamus may also be responsible for maintaining the rhythmicity of ictal discharges.

Intracranial stimulation for children with epilepsy.

OBJECTIVES: To evaluate the efficacy of intracranial stimulation to treat refractory epilepsy in children. METHODS: This is a retrospective analysis of a pilot study on all 8 children who had intracranial electrical stimulation for the investigation and treatment of refractory epilepsy at King's College Hospital between 2014 and 2015. Five children (one with temporal lobe epilepsy and four with frontal lobe epilepsy) had subacute cortical stimulation (SCS) for a period of 20-161 h during intracranial video-telemetry. Efficacy of stimulation was evaluated by counting interictal discharges and seizures. Two children had thalamic deep brain stimulation (DBS) of the centromedian nucleus (one with idiopathic generalized epilepsy, one with presumed symptomatic generalized epilepsy), and one child on the anterior nucleus (right fronto-temporal epilepsy). The incidence of interictal discharges was evaluated visually and quantified automatically. RESULTS: Among the three children with DBS, two had >60% improvement in seizure frequency and severity and one had no improvement. Among the five children with SCS, four showed improvement in seizure frequency (>50%) and one chid did not show improvement. Procedures were well tolerated by children. CONCLUSION: Cortical and thalamic stimulation appear to be effective and well tolerated in children with refractory epilepsy. SCS can be used to identify the focus and predict the effects of resective surgery or chronic cortical stimulation. Further larger studies are necessary.

Detection of Intracranial Signatures of Interictal Epileptiform Discharges from Concurrent Scalp EEG.

Interictal epileptiform discharges (IEDs) are transient neural electrical activities that occur in the brain of patients with epilepsy. A problem with the inspection of IEDs from the scalp electroencephalogram (sEEG) is that for a subset of epileptic patients, there are no visually discernible IEDs on the scalp, rendering the above procedures ineffective, both for detection purposes and algorithm evaluation. On the other hand, intracranially placed electrodes yield a much higher incidence of visible IEDs as compared to concurrent scalp electrodes. In this work, we utilize concurrent scalp and intracranial EEG (iEEG) from a group of temporal lobe epilepsy (TLE) patients with low number of scalp-visible IEDs. The aim is to determine whether by considering the timing information of the IEDs from iEEG, the resulting concurrent sEEG contains enough information for the IEDs to be reliably distinguished from non-IED segments. We develop an automatic detection algorithm which is tested in a leave-subject-out fashion, where each test subject's detection algorithm is based on the other patients' data. The algorithm obtained a [Formula: see text] accuracy in recognizing scalp IED from non-IED segments with [Formula: see text] accuracy when trained and tested on the same subject. Also, it was able to identify nonscalp-visible IED events for most patients with a low number of false positive detections. Our results represent a proof of concept that IED information for TLE patients is contained in scalp EEG even if they are not visually identifiable and also that between subject differences in the IED topology and shape are small enough such that a generic algorithm can be used.

Prognostic value of the second ictal intracranial pattern for the outcome of epilepsy surgery.

OBJECTIVE: To investigate the prognostic value of the second ictal pattern (SIP) that follows the first ictal pattern (FIP) seen at seizure onset in order to predict seizure control after epilepsy surgery. METHODS: SIPs were analysed in 344 electro-clinical and subclinical seizures recorded with intracranial electrodes in 63 patients. SIPs were classified as (a) electrodecremental event (EDE); (b) fast activity (FA); (c) runs of spikes; (d) spike-wave activity; (e) sharp waves; (f) alpha activity; (g) delta activity and (h) theta activity. Engel surgical outcome scale was used. RESULTS: The mean follow-up period was 42.1 months (SD=30.1). EDE was the most common SIP seen (41%), followed by FA (19%), spike-wave activity (18%), alpha activity (8%), sharp-wave activity (8%), delta activity (3%), runs of spikes (2%) and theta activity (2%). EDE as SIP was associated with favourable outcome when compared with FA (p=0.0044) whereas FA was associated with poor outcome when compared with any other pattern (p=0.0389). FA as SIP tends to occur after EDE (75%) whereas EDE tends to evolve from a FIP containing FA (77%). SIP extent was focal in 46% of patients, lobar in 24%, multilobar in 14% and bilateral in 16%. There is a gradual decrease in the proportion of Engel grade I with the extent of SIP. Focal and delayed (in temporal lobe epilepsy) SIPs appear to be associated with better outcome. CONCLUSIONS: As SIP, EDE was associated with favourable surgical outcome whereas FA was associated with poor outcome, probably because outcome is dominated by FIP. SIGNIFICANCE: EDE as SIP should not discourage surgery. However, FA as SIP should be contemplated with caution. SIP focality and latency can have prognostic value in epilepsy surgery.

Electrical Stimulation of the Anterior Cingulate Gyrus Induces Responses Similar to K-complexes in Awake Humans.

BACKGROUND: The brain region responsible for the initiation of K-complexes has not been identified to date. OBJECTIVE: To determine the brain region responsible for originating K-complexes. METHODS: We reviewed all 269 patients assessed for epilepsy surgery with intracranial electrodes and single pulse electrical stimulation (SPES) at King's College Hospital between 1999 and 2013. Intracranial EEG responses to electrical stimulation at orbitofrontal, frontal, cingulate, temporal and parietal loci were compared visually with each patient's K-complexes and the degree of resemblance was quantified. RESULTS: Among the 269 patients, K-complex-like responses were exclusively observed in all 6 patients who had depth electrodes in the cingulate cortex. In each patient, the stimulation site eliciting the response of greatest similarity to the patient's K-complex was located within the dorso-caudal anterior cingulate. The K-complex like responses were evoked when the patients were awake. CONCLUSION: Our findings provide the first causal evidence that the cingulate gyrus initiates the widespread synchronous activity that constitutes the K-complex. The induction of K-complex-like responses during wakefulness suggests that the mechanisms required for the initiation of K-complexes are separate from those involved in sleep.

Incidence of functional bi-temporal connections in the human brain in vivo and their relevance to epilepsy surgery.

The incidence of functional connections between human temporal lobes and their latencies were investigated using intracranial EEG responses to electrical stimulation with 1 msec single pulses in 91 patients assessed for surgery for treatment of epilepsy. The areas studied were amygdala, hippocampus, parahippocampal gyrus, fusiform gyrus, inferior and mid temporal gyrus. Furthermore, we assessed whether the presence of such connections are related to seizure onset extent and postsurgical seizure control. Responses were seen in any region of the contralateral temporal lobe when stimulating temporal regions in 30 patients out of the 91 (32.96%). Bi-hippocampal or bi-amygdalar projections were seen in only 5% of temporal lobes (N = 60) and between both fusiform gyri in 7.1% (N = 126). All other bilateral connections occurred in less than 5% of hemispheres. Depending on the structures, latencies ranged between 20 and 90 msec, with an average value of 60.2 msec. There were no statistical difference in the proportion of patients showing Engel Class I between patients with and without contralateral temporal connections. No difference was found in the proportion of patients showing bilateral or unilateral seizure onset among patients with and without contralateral temporal projections. The present findings corroborate that the functionality of bilateral temporal connections in humans is limited and does not affect the surgical outcome.

Dampened hippocampal oscillations and enhanced spindle activity in an asymptomatic model of developmental cortical malformations.

Developmental cortical malformations comprise a large spectrum of histopathological brain abnormalities and syndromes. Their genetic, developmental and clinical complexity suggests they should be better understood in terms of the complementary action of independently timed perturbations (i.e., the multiple-hit hypothesis). However, understanding the underlying biological processes remains puzzling. Here we induced developmental cortical malformations in offspring, after intraventricular injection of methylazoxymethanol (MAM) in utero in mice. We combined extensive histological and electrophysiological studies to characterize the model. We found that MAM injections at E14 and E15 induced a range of cortical and hippocampal malformations resembling histological alterations of specific genetic mutations and transplacental mitotoxic agent injections. However, in contrast to most of these models, intraventricularly MAM-injected mice remained asymptomatic and showed no clear epilepsy-related phenotype as tested in long-term chronic recordings and with pharmacological manipulations. Instead, they exhibited a non-specific reduction of hippocampal-related brain oscillations (mostly in CA1); including theta, gamma and HFOs; and enhanced thalamocortical spindle activity during non-REM sleep. These data suggest that developmental cortical malformations do not necessarily correlate with epileptiform activity. We propose that the intraventricular in utero MAM approach exhibiting a range of rhythmopathies is a suitable model for multiple-hit studies of associated neurological disorders.