Factors influencing the duration of generalized tonic–clonic seizure

Deep learning-based detection of generalized convulsive seizures using a wrist-worn accelerometer

OBJECTIVE

To develop and validate a wrist-worn accelerometer-based, deep-learning tunable algorithm for the automated detection of generalized or bilateral convulsive seizures (CSs) to be integrated with off-the-shelf smartwatches.

METHODS

We conducted a prospective multi-center study across eight European epilepsy monitoring units, collecting data from 384 patients undergoing video electroencephalography (vEEG) monitoring with a wrist-worn three dimensional (3D)-accelerometer sensor. We developed an ensemble-based convolutional neural network architecture with tunable sensitivity through quantile-based aggregation. The model, referred to as Episave, used accelerometer amplitude as input. It was trained on data from 37 patients who had 54 CSs and evaluated on an independent dataset comprising 347 patients, including 33 who had 49 CSs.

RESULTS

Cross-validation on the training set showed that optimal performance was obtained with an aggregation quantile of 60, with a 98% sensitivity, and a false alarm rate (FAR) of 1/6 days. Using this quantile on the independent test set, the model achieved a 96% sensitivity (95% confidence interval [CI]: 90%-100%), a FAR of <1/8 days (95% CI: 1/9-1/7 days) with 1 FA/61 nights, and a median detection latency of 26 s. One of the two missed CSs could be explained by the patient's arm, which was wearing the sensor, being trapped in the bed rail. Other quantiles provided up to 100% sensitivity at the cost of a greater FAR (1/2 days) or very low FAR (1/100 days) at the cost of lower sensitivity (86%).

SIGNIFICANCE

This Phase 2 clinical validation study suggests that deep learning techniques applied to single-sensor accelerometer data can achieve high CS detection performance while enabling tunable sensitivity.

Tonic and tonic-clonic seizures in the first year of life: Insights from electrographic features

OBJECTIVE

We studied the electrographic features of tonic seizures (TS) with bilateral contraction and tonic-clonic seizures (TCS) without focal signs occurring during the first year of life to evaluate if there is a correlation with outcome.

METHODS

We retrospectively reviewed patients aged 1 to 12 months with at least one TS or TCS recorded with video-EEG between 2011 and 2021 in our Epilepsy Monitoring Unit. We analyzed the following electrographic features: seizure duration, presence and duration of focal ictal EEG onset, and post-ictal generalized EEG suppression (PGES). Among clinical variables, we collected age at epilepsy onset, age at TS and TCS recording, response to anti-seizure medications, genetic and neuroimaging findings, epileptic syndrome classification.

RESULTS

Overall, we recorded 2577 seizures in 1769 patients. One-hundred-twenty-eight seizures (5%) were clinically labeled either as TS or TCS in 41 patients (2%). Out of 41 patients, 17 (41%) presented with TS, and 24 (59%) with TCS. Thirteen patients (32%) had a Self-limited Epilepsy, and 28 (68%) a Developmental and Epileptic Encephalopathy (DEE). Seventy-two percent of genetically tested patients had pathogenic gene variants. None had structural epilepsy. Mean age at epilepsy onset was 4.48 months (range 3 days-12 months). Age at seizure onset was earlier in patients presenting with TS versus patients presenting with TCS (2.31 months vs. 6.01 months; p = 0.001) and in DEEs versus Self-limited Epilepsies (3.23 months vs. 7.16 months; p = 0.001). TS were exclusively present in DEEs (p = 0.001), and TCS were recorded in both DEEs and Self-limited Epilepsies. Focal ictal EEG onset was evident in 92 % of TCS, and in none of TS. Generalized ictal EEG onset was documented in 100 % of TS, and in 8 % of TCS. Focal ictal EEG onset occurred more frequently (100 % vs. 32 %; p = 0.000) and was significantly longer (30.61 s vs. 16.22 s; p = 0.020) in Self-limited Epilepsies versus DEEs. PGES was observed in 18 out of 41 (44 %) and was more frequent in Self-limited Epilepsies (p = 0.026).

SIGNIFICANCE

This study provides insights into the electroclinical features of TS and TCS in infants that may help distinguish Self-limited Epilepsies from DEEs soon after epilepsy onset.

Prospective multicenter study of continuous tonic-clonic seizure monitoring on Apple Watch in epilepsy monitoring units and ambulatory environments

OBJECTIVE

Evaluate the performance of a custom application developed for tonic-clonic seizure (TCS) monitoring on a consumer-wearable (Apple Watch) device.

METHODS

Participants with a history of convulsive epileptic seizures were recruited for either Epilepsy Monitoring Unit (EMU) or ambulatory (AMB) monitoring; participants without epilepsy (normal controls [NC]) were also enrolled in the AMB group. Both EMU and AMB participants wore an Apple Watch with a research app that continuously recorded accelerometer and photoplethysmography (PPG) signals, and ran a fixed-and-frozen tonic-clonic seizure detection algorithm during the testing period. This algorithm had been previously developed and validated using a separate training dataset. All EMU convulsive events were validated by video-electroencephalography (video-EEG); AMB events were validated by caregiver reporting and follow-ups. Device performance was characterized and compared to prior monitoring devices through sensitivity, false alarm rate (FAR; false-alarms per 24 h), precision, and detection delay (latency).

RESULTS

The EMU group had 85 participants (4,279 h, 19 TCS from 15 participants) enrolled across four EMUs; the AMB group had 21 participants (13 outpatient, 8 NC, 6,735 h, 10 TCS from 3 participants). All but one AMB participant completed the study. Device performance in the EMU group included a sensitivity of 100 % [95 % confidence interval (CI) 79-100 %]; an FAR of 0.05 [0.02, 0.08] per 24 h; a precision of 68 % [48 %, 83 %]; and a latency of 32.07 s [standard deviation (std) 10.22 s]. The AMB group had a sensitivity of 100 % [66-100 %]; an FAR of 0.13 [0.08, 0.24] per 24 h; a precision of 22 % [11 %, 37 %]; and a latency of 37.38 s [13.24 s]. Notably, a single AMB participant was responsible for 8 of 31 false alarms. The AMB FAR excluding this participant was 0.10 [0.07, 0.14] per 24 h.

DISCUSSION

This study demonstrates the practicability of TCS monitoring on a popular consumer wearable (Apple Watch) in daily use for people with epilepsy. The monitoring app had a high sensitivity and a substantially lower FAR than previously reported in both EMU and AMB environments.

Distinctive Clinico-electrographic and Radiological Profile of Childhood and Adolescent Seizures

AIM

Electroencephalogram (EEG) is specific, but not sensitive, for the diagnosis of epilepsy. This study aimed to correlate the clinico-electrographic and radiological features of seizure disorders in children attending a tertiary care centre in northern India.

METHODS

Children aged between one to 18 years with seizure episodes were included. Clinical details, including historical as well as physical findings, were evaluated along with EEG and neuroimaging (Magnetic resonance imaging). Details were noted on pre-designed proforma. Variables were analysed by using appropriate statistical methods.

RESULTS

A total of 110 children with seizures were enrolled in the study. Male to female ratio was 1.6: 1, and the mean age of the study children was 8 years. The majority of the children were symptomatic for more than one year. The most common seizure type was Generalised Tonic Clonic Seizure (GTCS), and Hypoxic-ischemic Encephalopathy (HIE) sequelae was the most commonly attributed etiology, followed by neurocysticercosis. EEG and neuroimaging findings were found to correlate well with seizure semiology from history. The incidence of febrile seizures was 10% in this study, with nearly three-fourths of them being simple febrile seizures.

CONCLUSION

Microcephaly and developmental delay were the most distinctive clinical correlates in children with seizures. There was a fair agreement between the types of seizures described in history and depicted on EEG with Cohen's kappa of 0.4. Also, there was a significant association between the type of seizures on EEG and the duration of symptoms.

Prospective Study of a Multimodal Convulsive Seizure Detection Wearable System on Pediatric and Adult Patients in the Epilepsy Monitoring Unit

Background: Using machine learning to combine wrist accelerometer (ACM) and electrodermal activity (EDA) has been shown effective to detect primarily and secondarily generalized tonic-clonic seizures, here termed as convulsive seizures (CS). A prospective study was conducted for the FDA clearance of an ACM and EDA-based CS-detection device based on a predefined machine learning algorithm. Here we present its performance on pediatric and adult patients in epilepsy monitoring units (EMUs).

Methods: Patients diagnosed with epilepsy participated in a prospective multi-center clinical study. Three board-certified neurologists independently labeled CS from video-EEG. The Detection Algorithm was evaluated in terms of Sensitivity and false alarm rate per 24 h-worn (FAR) on all the data and on only periods of rest. Performance were analyzed also applying the Detection Algorithm offline, with a less sensitive but more specific parameters configuration (“Active mode”).

Results: Data from 152 patients (429 days) were used for performance evaluation (85 pediatric aged 6–20 years, and 67 adult aged 21–63 years). Thirty-six patients (18 pediatric) experienced a total of 66 CS (35 pediatric). The Sensitivity (corrected for clustered data) was 0.92, with a 95% confidence interval (CI) of [0.85-1.00] for the pediatric population, not significantly different (p > 0.05) from the adult population's Sensitivity (0.94, CI: [0.89–1.00]). The FAR on the pediatric population was 1.26 (CI: [0.87–1.73]), higher (p < 0.001) than in the adult population (0.57, CI: [0.36–0.81]). Using the Active mode, the FAR decreased by 68% while reducing Sensitivity to 0.95 across the population. During rest periods, the FAR's were 0 for all patients, lower than during activity periods (p < 0.001).

Conclusions: Performance complies with FDA's requirements of a lower bound of CI for Sensitivity higher than 0.7 and of a FAR lower than 2, for both age groups. The pediatric FAR was higher than the adult FAR, likely due to higher pediatric activity. The high Sensitivity and precision (having no false alarms) during sleep might help mitigate SUDEP risk by summoning caregiver intervention. The Active mode may be advantageous for some patients, reducing the impact of the FAR on daily life. Future work will examine the performance and usability outside of EMUs.

Postictal generalized EEG suppression and respiratory dysfunction following generalized tonic-clonic seizures in sleep and wakefulness

OBJECTIVE

Sudden unexpected death in epilepsy (SUDEP) is a common cause of death in epilepsy and frequently occurs following generalized tonic-clonic seizures (GTCS) in sleep. Postictal generalized electroencephalography (EEG) suppression (PGES), postictal immobility, and periictal respiratory dysfunction are potential risk factors for SUDEP. We sought to determine whether there was a difference in respiratory dysfunction, PGES, and postictal immobility for GTCS occurring during wakefulness or sleep.

METHODS

We retrospectively analyzed video-EEG telemetry data in the epilepsy-monitoring unit. Patients' state at seizure onset and seizure characteristics were identified. Respiratory parameters and heart rate were recorded. Presence and duration of PGES and time to first postictal nonrespiratory movement were recorded.

RESULTS

There were 165 seizures in 67 patients. There was no significant difference in the duration of postictal immobility in GTCS occurring out of wakefulness or sleep (p = 0.280). Oxygen desaturation nadir (p = 0.572) and duration of oxygen desaturation were not significantly different for GTCS starting during sleep or wakefulness (p = 0.992). PGES occurred more frequently when seizure onset was in sleep than in wakefulness (p = 0.004; odds ratio [OR] 2.760). There was no difference in the duration of PGES between the two groups.

SIGNIFICANCE

PGES occurs more commonly after GTCS in sleep than in wakefulness but, in the epilepsy-monitoring unit (EMU), a patient's state at seizure onset does not affect the degree of respiratory dysfunction or duration of postictal immobility. In sleep, outside the hospital setting, GTCS are likely to go unnoticed. Postictal immobility in prone patients prevents head repositioning and unimpeded air exchange. A positive feedback cycle ensues with increasing respiratory distress, potentiating postictal immobility and PGES and eventually leading to asystole. Our findings suggest that the high incidence of nocturnal SUDEP may be related to the unsupervised environment during sleep rather than the severity of sleep-related respiratory dysfunction or PGES duration in the immediate postictal period.

Paroxysmal events during prolonged video-video electroencephalography monitoring in refractory epilepsy

INTRODUCTION

Given that epileptic seizures and non-epileptic paroxysmal events have similar clinical manifestations, using specific diagnostic methods is crucial, especially in patients with drug-resistant epilepsy. Prolonged video electroencephalography monitoring during epileptic seizures reveals epileptiform discharges and has become an essential procedure for epilepsy diagnosis. The main purpose of this study is to characterise paroxysmal events and compare patterns in patients with refractory epilepsy.

METHODS

We conducted a retrospective analysis of medical records from 91 patients diagnosed with refractory epilepsy who underwent prolonged video electroencephalography monitoring during hospitalisation.

RESULTS

During prolonged video electroencephalography monitoring, 76.9% of the patients (n=70) had paroxysmal events. The mean number of events was 3.4±2.7; the duration of these events was highly variable. Most patients (80%) experienced seizures during wakefulness. The most common events were focal seizures with altered levels of consciousness, progressive bilateral generalized seizures and psychogenic non-epileptic seizures. Regarding all paroxysmal events, no differences were observed in the number or type of events by sex, in duration by sex or age at onset, or in the number of events by type of event. Psychogenic nonepileptic seizures were predominantly registered during wakefulness, lasted longer, started at older ages, and were more frequent in women.

CONCLUSIONS

Paroxysmal events recorded during prolonged video electroencephalography monitoring in patients with refractory epilepsy show similar patterns and characteristics to those reported in other latitudes.