Time to FIRE NORSE: A single acronym for a heterogeneous presentation. Further information from a case series and discussion of the literature.

PURPOSE: New-Onset Refractory Status Epilepticus (NORSE) is a rare and severe form of refractory status epilepticus without an apparent underlying cause at presentation or prior history of epilepsy. We aimed to describe the clinical features, etiology, treatment, and outcomes of NORSE in adults in a quaternary-level hospital in Saudi Arabia. METHODOLOGY: In this retrospective cohort study, inclusion criteria involved patients over 14 years old who met the 2018 consensus definition for NORSE. Patients were identified using a combination of medical record admission labels 'status epilepticus' and 'encephalitis', and continuous EEG reports documenting status epilepticus. Demographic, clinical, and radiological data were collected and then analyzed for factors correlated with specific etiologies, better functional outcomes, and future diagnosis of epilepsy. RESULTS: We found 24 patients presenting with NORSE between 2010 and 2021. Fever/infectious symptoms were the most common prodrome. Elevated inflammatory serum and cerebrospinal fluid markers in most patients. Brain MRI revealed T2/FLAIR hyperintensity patterns, predominantly affecting limbic and perisylvian structures. The etiology of NORSE varied, with immune-related causes being the most common. Long-term outcomes were poor, with a high mortality rate and most survivors developing drug-resistant epilepsy. CONCLUSION: This study provides valuable insights into NORSE's clinical characteristics, highlighting the heterogeneity of this condition. The poor outcome is likely related to the progressive nature of the underlying disease, where refractory seizures are a clinical symptom. Thus, we propose to focus future research on the etiology rather than the NORSE acronym.

Dynamic interactions between anterior insula and anterior cingulate cortex link perceptual features and heart rate variability during movie viewing.

The dynamic integration of sensory and bodily signals is central to adaptive behaviour. Although the anterior cingulate cortex (ACC) and the anterior insular cortex (AIC) play key roles in this process, their context-dependent dynamic interactions remain unclear. Here, we studied the spectral features and interplay of these two brain regions using high-fidelity intracranial-EEG recordings from five patients (ACC: 13 contacts, AIC: 14 contacts) acquired during movie viewing with validation analyses performed on an independent resting intracranial-EEG dataset. ACC and AIC both showed a power peak and positive functional connectivity in the gamma (30-35 Hz) frequency while this power peak was absent in the resting data. We then used a neurobiologically informed computational model investigating dynamic effective connectivity asking how it linked to the movie's perceptual (visual, audio) features and the viewer's heart rate variability (HRV). Exteroceptive features related to effective connectivity of ACC highlighting its crucial role in processing ongoing sensory information. AIC connectivity was related to HRV and audio emphasising its core role in dynamically linking sensory and bodily signals. Our findings provide new evidence for complementary, yet dissociable, roles of neural dynamics between the ACC and the AIC in supporting brain-body interactions during an emotional experience.

Resource use and costs associated with epilepsy in the Queensland hospital system: protocol for a population-based data linkage study.

INTRODUCTION: Epilepsy places a large burden on health systems, with hospitalisations for seizures alone occurring more frequently than those related to diabetes. However, the cost of epilepsy to the Australian health system is not well understood. The primary aim of this study is to quantify the health service use and cost of epilepsy in Queensland, Australia. Secondary aims are to identify differences in health service use and cost across population and disease subgroups, and to explore the associations between health service use and common comorbidities. METHODS AND ANALYSIS: This project will use data linkage to identify the health service utilisation and costs associated with epilepsy. A base cohort of patients will be identified from the Queensland Hospital Admitted Patient Data Collection. We will select all patients admitted between 2014 and 2018 with a diagnosis classification related to epilepsy. Two comparison cohorts will also be identified. Retrospective hospital admissions data will be linked with emergency department presentations, clinical costing data, specialist outpatient and allied health occasions of service data and mortality data. The level of health service use in Queensland, and costs associated with this, will be quantified using descriptive statistics. Difference in health service costs between groups will be explored using logistic regression. Linear regression will be used to model the associations of interest. The analysis will adjust for confounders including age, sex, comorbidities, indigenous status, and remoteness. ETHICS AND DISSEMINATION: Ethical approval has been obtained through the QUT University Human Research Ethics Committee (1900000333). Permission to waive consent has been granted under the Public Health Act 2005, with approval provided by all relevant data custodians. Findings of the proposed research will be communicated through presentations at national and international conferences, presentations to key stakeholders and decision-makers, and publications in international peer-reviewed journals.

Psychogenic nonepileptic seizures treated as epileptic seizures in the emergency department.

OBJECTIVES: We aimed to estimate the rate of psychogenic nonepileptic seizures (PNES) among patients presenting to an emergency department with presumed seizures. We also wanted to identify factors that can assist health care professionals in determining whether these events are likely to be epileptic or nonepileptic. METHODS: We performed two retrospective audits on patients who were treated for seizures in the department of emergency medicine at the Princess Alexandra Hospital, Brisbane, Australia. Exploratory analyses and logistic regressions were conducted to investigate the characteristics of the presentations and the relationships between our variables of interest. RESULTS: In the group of all presentations with presumed seizures over a 3-month period (n = 157), a total of 151 presentations (96.2%) presentations were given a primary diagnosis of epileptic seizures. Of these 151 presentations, only 84 (55.6%) presented with epileptic seizures and 40 (26.5%) actually presented with PNES. In the group of patients who presented with prolonged and/or multiple events (n = 213) over a 1-year period, 196 (92.0%) were treated as epileptic seizures. Of these 196 presentations, only 85 (43.4%) presented with epileptic seizures and 97 (49.5%) actually presented with PNES. Several factors were identified to help risk stratify between epileptic seizures and PNES: Duration of events and of the postictal phase, number of events, presence of a structural brain pathology, mental health history, lactate levels and presence of tongue bite, incontinence, and/or vomiting. SIGNIFICANCE: A large proportion of people who present to emergency departments with events resembling epileptic seizures actually have PNES rather than epilepsy-particularly those patients who present with prolonged and/or multiple events. The rate of misdiagnosis was high. Efforts need to be made to recognize patients with psychogenic nonepileptic seizures earlier and diagnose them correctly to avoid unnecessary iatrogenic harm and to provide adequate treatment.

Ictal fear is associated with anxiety symptoms and interictal dysphoric disorder in drug-resistant mesial temporal lobe epilepsy.

Interictal dysphoric disorder (IDD) is a poorly understood psychiatric disorder of epilepsy patients. Interictal dysphoric disorder is characterized by depressive, somatoform, and affective symptoms observed in up to 5.9% of drug-resistant mesial temporal lobe epilepsy with hippocampal sclerosis (MTLE-HS). This study aimed to evaluate the association between ictal fear (IF) and the psychiatric symptoms and diagnosis in MTLE-HS patients. We included 116 (54.3% male) consecutive adult patients (36 ±â€¯11 years) with MTLE-HS. Anxiety and depression symptoms were evaluated by the Hospital Anxiety and Depression Scale (HADS) and the psychiatric diagnosis were according to Fourth Edition of the Diagnosis and Statistical Manual of Mental Disorders (DSM-IV). The independent association between the occurrence of IF aura and the psychiatric diagnosis was determined by binary regression. When compared to those with other auras or without aura, patients reporting IF have higher HADS anxiety, but not HADS depression, scores. Ictal fear was independently associated with the diagnosis of interictal dysphoric disorder (OR, IC 95% = 7.6, 1.3-43.2, p = 0.02), but not with the diagnosis of anxiety (OR, CI 95% = 0.72, 0.08-6.0, p = 0.73), depression (OR, CI 95% = 0.94, 0.19-4.8, p = 0.94) or psychotic disorders (p = 0.99). Only patients with drug-resistant MTLE-HS were included and the small number of cases with DD diagnosis in the sample. In MTLE-HS patients, the occurrence of IF is associated with higher levels of anxiety symptoms and IDD. The results provide insights about fear-related neural network connections with anxiety symptoms and the IDD in MTLE-HS.

Intracranial-EEG evidence for medial temporal pole driving amygdala activity induced by multi-modal emotional stimuli.

The temporal pole (TP) is an associative cortical region required for complex cognitive functions such as social and emotional cognition. However, mapping the TP with functional magnetic resonance imaging is technically challenging and thus understanding its interaction with other key emotional circuitry, such as the amygdala, remains elusive. We exploited the unique advantages of stereo-electroencephalography (sEEG) to assess the responses of the TP and the amygdala during the perception of emotionally salient stimuli of pictures, music and movies. These stimuli consistently elicited high gamma responses (70-140 Hz) in both the TP and the amygdala, accompanied by functional connectivity in the low frequency range (2-12 Hz). Computational analyses suggested that the TP drove this effect in the theta frequency range, modulated by the emotional valence of the stimuli. Notably, cross-frequency analysis indicated the phase of theta oscillations in the TP modulated the amplitude of high gamma activity in the amygdala. These results were reproducible across three types of sensory inputs including naturalistic stimuli. Our results suggest that multimodal emotional stimuli induce a hierarchical influence of the TP over the amygdala.

Quantifying volume conducted potential using stimulation artefact in cortico-cortical evoked potentials.

BACKGROUND: Cortico-cortical evoked potentials (CCEP) are a technique using low frequency stimulation to infer regions of cortical connectivity in patients undergoing Stereo-electroencephalographic (SEEG) monitoring for refractory epilepsy. Little attention has been given to volume conducted components of CCEP responses, and how they may inflate CCEP connectivity. NEW METHOD: Using data from 37 SEEG-CCEPs patients, a novel method was developed to quantify stimulation artefact by measuring the peak-to-peak voltage difference in the first 10 ms after CCEP stimulation. Early responses to CCEP stimulation were also quantified by calculating the root mean square of the 10-100 ms period after each stimulation pulse. Both the early CCEP responses and amplitude of stimulation artefact were regressed by physical distance, stimulation waveform, stimulation intensity and tissue type to identify conduction related properties. RESULTS: Both stimulation artefact and early responses were correlated strongly with the inverse square of the distance from the stimulating electrode. Once corrected for the inverse square distance from the electrode, stimulation artefact and CCEP responses showed a linear relationship, indicating a volume conducted component. COMPARISON WITH EXISTING METHODS: This is the first study to use stimulation artefact to quantify volume conducted potentials, and is the first to quantify volume conducted potentials in SEEG. A single prior study utilizing electrocorticography has shown that parts of early CCEP responses are due to volume conduction. CONCLUSIONS: The linear relationship between stimulation artefact amplitude and CCEP early responses, once corrected for distance, suggests that stimulation artefact can be used as a measure to quantify the volume conducted components.

Comparing connectivity metrics in cortico-cortical evoked potentials using synthetic cortical response patterns.

BACKGROUND: Cortico-Cortical Evoked Potentials (CCEPs) are a novel low frequency stimulation method used for brain mapping during intracranial epilepsy investigations. Only a handful of metrics have been applied to CCEP data to infer connectivity, and no comparison as to which is best has been performed. NEW METHOD: We implement a novel method which involved superimposing synthetic cortical responses onto stereoelectroencephalographic (SEEG) data, and use this to compare several metric's ability to detect the simulated patterns. In this we compare two commonly employed metrics currently used in CCEP analysis against eight time series similarity metrics (TSSMs), which have been widely used in machine learning and pattern matching applications. RESULTS: Root Mean Square (RMS), a metric commonly employed in CCEP analysis, was sensitive to a wide variety of response patterns, but insensitive to simulated epileptiform patterns. Autoregressive (AR) coefficients calculated by Burg's method were also sensitive to a wide range of patterns, but were extremely sensitive to epileptiform patterns. Other metrics which employed elastic warping techniques were less sensitive to the simulated response patterns. COMPARISON WITH EXISTING METHODS: Our study is the first to compare CCEP connectivity metrics against one-another. Our results found that RMS, which has been used in many CCEP studies previously, was the most sensitive metric across a wide range of patterns. CONCLUSIONS: Our novel method showed that RMS is a robust and sensitive measure, validating much of the findings of the SEEG-CCEP literature to date. Autoregressive coefficients may also be a useful metric to investigate epileptic networks.

Intracranial EEG recording in autoimmune epilepsy dataset.

The appearance of seizures in autoimmune epilepsy on intracranial recordings has not been previously demonstrated. The following data shows a multifocal epilepsy in a patient with seronegative autoimmune epilepsy (reported here; "Electroclinical Insights into Autoimmune Epilepsy", Gillinder, 2019). Independent seizures were seen to arise from 5 separate foci. These all began with slow repetitive spiking in a highly restricted area. Only after many minutes would this activity spread to other regions. Despite arising from different locations, all foci affected the posterior insula resulting in clinical symptoms.

Connectivity of the human insula: A cortico-cortical evoked potential (CCEP) study.

OBJECTIVE: The human insula is increasingly being implicated as a multimodal functional network hub involved in a large variety of complex functions. Due to its inconspicuous location and highly vascular anatomy, it has historically been difficult to study. Cortico-cortical evoked potentials (CCEPs), utilize low frequency stimulation to map cerebral networks. They were used to study connections of the human insula. METHODS: CCEP data was acquired from each sub-region of the dominant and non-dominant insula in 30 patients who underwent stereo-EEG. Connectivity strength to the various cortical regions was obtained via a measure of root mean square (RMS), calculated from each gyrus of the insula and ranked into weighted means. RESULTS: The results of all cumulative CCEP responses for each individual gyrus were represented by circro plots. Forty-nine individual CCEP pairs were stimulated across all the gyri from the right and left insula. In brief, the left insula contributed more greatly to language areas. Sensory function, pain, saliency processing and vestibular function were more heavily implicated from the right insula. Connections to the primary auditory cortex arose from both insula regions. Both posterior insula regions showed significant contralateral connectivity. Ipsilateral mesial temporal connections were seen from both insula regions. In visual function, we further report the novel finding of a direct connection between the right posterior insula and left visual cortex. SIGNIFICANCE: The insula is a major multi-modal network hub with the cerebral cortex having major roles in language, sensation, auditory, visual, limbic and vestibular functions as well as saliency processing. In temporal lobe epilepsy surgery failure, the insula may be implicated as an extra temporal cause, due to the strong mesial temporal connectivity findings.

Automatic detection of the epileptogenic zone: An application of the fingerprint of epilepsy.

BACKGROUND: The successful delineation of the epileptogenic zone in epilepsy monitoring is crucial for achieving seizure freedom after epilepsy surgery. NEW METHOD: We aim to improve epileptogenic zone localization by utilizing a computer-assisted tool for the automated grading of the seizure activity recorded in various locations for 20 patients undergoing stereo electroencephalography. Their epileptic seizures were processed to extract two potential biomarkers. The concentration of these biomarkers from within each patient's implantation were then graded to identify their epileptogenic zone and were compared to the clinical assessment. RESULTS: Our technique was capable of ranking the clinically defined epileptogenic zone with high accuracy, above 95%, with a true to false positive ratio of 1:1.52, and was effective with both temporal and extra-temporal onset epilepsies. COMPARISON WITH EXISTING METHOD: We compared our method to two other groups performing localization using similar biomarkers. Our classification metrics, sensitivity and precision together were comparable to both groups and our overall accuracy from a larger population was also higher then both. CONCLUSIONS: Our method is highly accurate, automated and non-parametric providing clinicians another tool that can be used to help identify the epileptogenic zone in patients undergoing the stereo electroencephalography procedure for epilepsy monitoring.

Nocturnal motor events in epilepsy: Is there a defined physiological network?

OBJECTIVE: Paroxysmal nocturnal movements in epilepsy are a recognised phenomenon, however, the mechanisms that produce them and the effect of the underlying epilepsy still remains elusive. In this study, 10 patients were studied to define the cerebral networks corresponding to these movements and explore how epileptiform activity modulated them. METHODS: We compared the change in power of the 25-250 Hz frequency band using event-related synchronization of all stereo-EEG electrodes implanted, during a baseline segment, during nocturnal movements and seizures. RESULTS: The underlying network activated during these paroxysmal movements comprised the insula, anterior cingulate, premotor areas and orbitofrontal regions. Three groups emerged, (1) complete overlap, (2) no overlap and (3) partial overlap of ERS changes of the epileptogenic zone within the proposed network and correlation of semiology between nocturnal movements and seizures. CONCLUSION: We conclude that nocturnal movements are due to a complex interplay within this physiological network of defined anatomical regions. Epileptic activity had significant impact on nocturnal movements but was not required for generation. SIGNIFICANCE: Where the semiology of the first clinical sign of a seizure consistently matches a patient's nocturnal movements, we suggest that the underlying epileptogenic zone is potentially located within this defined network.

Short burst Clobazam dosing at discharge from VEEG evaluation reduces re-presentation with seizures.

PURPOSE: Recurrent seizures and status epilepticus after medication reduction for inpatient Video Electroencephalograph (VEEG) monitoring is a well-known complication of this investigation. In the literature this is reported to occur at a rate of approximately 3-7%. We review the use of short burst Clobazam dosing on discharge from the Epilepsy monitoring unit (EMU) to determine if this might reduce rates of representation with seizures. METHODS: We performed a retrospective review of all cases admitted to the EMU. Their medication reduction, number of seizures, seizure severity and demographics were collected. Representations to hospital were considered if they occurred within 14 days of discharge from the unit. RESULTS: 264 cases were included, and 146 patients received 5 days of Clobazam 10 mg PO BD upon discharge after VEEG and 118 did not. There were significantly fewer patients re-presenting to hospital for seizures in the 14 days following discharge in those who were administered short-burst Clobazam compared to those who were not (0% and 4.23% respectively). There was also a trend towards fewer re-admissions for non-seizure indications including mental health issues or non-epileptic seizures and AED side effects. There were no definite adverse reactions to Clobazam recorded. CONCLUSION: Short burst Clobazam appears to be a safe and effective means to reduce representation with seizures after medication reduction during VEEG recording. This obviously benefits patients but it may also be a cost-effective means to reduce unnecessary health expenditure.

Electroclinical insights into autoimmune epilepsy.

PURPOSE: Chronic autoimmune epilepsy is an increasingly recognised entity however its clinical and electrographic features remain poorly understood. We present a case undergoing diagnostic Stereo-electroencephalography implantation that was found to have a multifocal perisylvian epilepsy with unique electrographic features and is now seizure free with immunotherapy. METHODS: The patient had antibody negative refractory perisylvian epilepsy and underwent implantation of the perisylvian-temporal networks. Immunomodulatory treatment was administered during SEEG. RESULTS: SEEG demonstrated a multifocal perisylvian epilepsy with strong involvement of the posterior insula. There was almost continuous spiking seen interictally from multiple foci within the right hemisphere and independent seizures were generated from 5 locations. After treatment with intravenous methylprednisone and immunoglobulin during SEEG, spiking and seizures terminated while still off anti-seizure medications. The patient remains seizure free on immunotherapy. CONCLUSION: This case highlights the importance of considering autoimmunity in the differential diagnosis of refractory epilepsy, especially perisylvian epilepsy. It also highlights the need to define a clinical phenotype associated with autoantibodies in epilepsy, as there are likely many cases who are not positive for one of the commercially available tests. This case also provides insights into the possible features of an electroclinical syndrome associated with autoimmunity.

Understanding Patients' Behavior: Vision-Based Analysis of Seizure Disorders.

A substantial proportion of patients with functional neurological disorders (FND) are being incorrectly diagnosed with epilepsy because their semiology resembles that of epileptic seizures (ES). Misdiagnosis may lead to unnecessary treatment and its associated complications. Diagnostic errors often result from an overreliance on specific clinical features. Furthermore, the lack of electrophysiological changes in patients with FND can also be seen in some forms of epilepsy, making diagnosis extremely challenging. Therefore, understanding semiology is an essential step for differentiating between ES and FND. Existing sensor-based and marker-based systems require physical contact with the body and are vulnerable to clinical situations such as patient positions, illumination changes, and motion discontinuities. Computer vision and deep learning are advancing to overcome these limitations encountered in the assessment of diseases and patient monitoring; however, they have not been investigated for seizure disorder scenarios. Here, we propose and compare two marker-free deep learning models, a landmark-based and a region-based model, both of which are capable of distinguishing between seizures from video recordings. We quantify semiology by using either a fusion of reference points and flow fields, or through the complete analysis of the body. Average leave-one-subject-out cross-validation accuracies for the landmark-based and region-based approaches of 68.1% and 79.6% in our dataset collected from 35 patients, reveal the benefit of video analytics to support automated identification of semiology in the challenging conditions of a hospital setting.

EEG findings in NMDA encephalitis - A systematic review.

PURPOSE: The clinical utility of EEG in cases of NMDA encephalitis is broad with many findings indicating not just epileptiform activity but also encephalopathy and potentially providing insights into pathophysiologic mechanisms of disease. We aimed to determine the frequency of different abnormalities described on EEG and their association with outcome in patients affected by NMDARE through a systematic review of all cases published. METHOD: A systematic literature review of PubMed and Embase of all published cases of anti-NMDA receptor encephalitis with EEG results, was performed from inception to January 2018. RESULTS: A total of 446 cases of anti-NMDA receptor encephalitis with reported EEG findings were identified. 373 EEGs were abnormal, and this strongly correlated with ICU admission and time to recovery (p = 0.014 and 0.04 respectively). ICU admission and recovery were also correlated with delta range abnormalities including extreme delta brush (p = 0.007 and 0.03). Electrographic seizures correlated strongly with clinical seizures (p < 0.0001), however only 39 cases had EEG seizures captured, while there were 294 cases with clinical seizures. CONCLUSIONS: EEG is useful in the clinical management and prognostication of cases on NMDA encephalitis. This is particularly true of certain findings which portend a higher likelihood of ICU admission or poorer outcome and this may assist in the decision to pursue more aggressive treatment options.

Aberrant epileptic seizure identification: A computer vision perspective.

PURPOSE: The recent explosion of artificial intelligence techniques in video analytics has highlighted the clinical relevance in capturing and quantifying semiology during epileptic seizures; however, we lack an automated anomaly identification system for aberrant behaviors. In this paper, we describe a novel system that is trained with known clinical manifestations from patients with mesial temporal and extra-temporal lobe epilepsy and presents aberrant semiology to physicians. METHODS: We propose a simple end-to-end-architecture based on convolutional and recurrent neural networks to extract spatiotemporal representations and to create motion capture libraries from 119 seizures of 28 patients. The cosine similarity distance between a test representation and the libraries from five aberrant seizures separate to the main dataset is subsequently used to identify test seizures with unusual patterns that do not conform to known behavior. RESULTS: Cross-validation evaluations are performed to validate the quantification of motion features and to demonstrate the robustness of the motion capture libraries for identifying epilepsy types. The system to identify unusual epileptic seizures successfully detects out of the five seizures categorized as aberrant cases. CONCLUSIONS: The proposed approach is capable of modeling clinical manifestations of known behaviors in natural clinical settings, and effectively identify aberrant seizures using a simple strategy based on motion capture libraries of spatiotemporal representations and similarities between hidden states. Detecting anomalies is essential to alert clinicians to the occurrence of unusual events, and we show how this can be achieved using pre-learned database of semiology stored in health records.

Vision-Based Mouth Motion Analysis in Epilepsy: A 3D Perspective.

Epilepsy monitoring involves the study of videos to assess clinical signs (semiology) to assist with the diagnosis of seizures. Recent advances in the application of vision-based approaches to epilepsy analysis have demonstrated significant potential to automate this assessment. Nevertheless, current proposed computer vision based techniques are unable to accurately quantify specific facial modifications, e.g. mouth motions, which are examined by neurologists to distinguish between seizure types. 2D approaches that analyse facial landmarks have been proposed to quantify mouth motions, however, they are unable to fully represent motions in the mouth and cheeks (ictal pouting) due to a lack of landmarks in the the cheek regions. Additionally, 2D region-based techniques based on the detection of the mouth have limitations when dealing with large pose variations, and thus make a fair comparison between samples difficult due to the variety of poses present. 3D approaches, on the other hand, retain rich information about the shape and appearance of faces, simplifying alignment for comparison between sequences. In this paper, we propose a novel network method based on a 3D reconstruction of the face and deep learning to detect and quantify mouth semiology in our video dataset of 20 seizures, recorded from patients with mesial temporal and extra-temporal lobe epilepsy. The proposed network is capable of distinguishing between seizures of both types of epilepsy. An average classification accuracy of 89% demonstrates the benefits of computer vision and deep learning for clinical applications of non-contact systems to identify semiology commonly encountered in a natural clinical setting.

Motion Signatures for the Analysis of Seizure Evolution in Epilepsy.

In epilepsy, semiology refers to the study of patient behavior and movement, and their temporal evolution during epileptic seizures. Understanding semiology provides clues to the cerebral networks underpinning the epileptic episode and is a vital resource in the pre-surgical evaluation. Recent advances in video analytics have been helpful in capturing and quantifying epileptic seizures. Nevertheless, the automated representation of the evolution of semiology, as examined by neurologists, has not been appropriately investigated. From initial seizure symptoms until seizure termination, motion patterns of isolated clinical manifestations vary over time. Furthermore, epileptic seizures frequently evolve from one clinical manifestation to another, and their understanding cannot be overlooked during a presurgery evaluation. Here, we propose a system capable of computing motion signatures from videos of face and hand semiology to provide quantitative information on the motion, and the correlation between motions. Each signature is derived from a sparse saliency representation established by the magnitude of the optical flow field. The developed computer-aided tool provides a novel approach for physicians to analyze semiology as a flow of signals without interfering in the healthcare environment. We detect and quantify semiology using detectors based on deep learning and via a novel signature scheme, which is independent of the amount of data and seizure differences. The system reinforces the benefits of computer vision for non-obstructive clinical applications to quantify epileptic seizures recorded in real-life healthcare conditions.