Artificial intelligence-enhanced epileptic seizure detection by wearables.

OBJECTIVE: Wrist- or ankle-worn devices are less intrusive than the widely used electroencephalographic (EEG) systems for monitoring epileptic seizures. Using custom-developed deep-learning seizure detection models, we demonstrate the detection of a broad range of seizure types by wearable signals. METHODS: Patients admitted to the epilepsy monitoring unit were enrolled and asked to wear wearable sensors on either wrists or ankles. We collected patients' electrodermal activity, accelerometry (ACC), and photoplethysmography, from which blood volume pulse (BVP) is derived. Board-certified epileptologists determined seizure onset, offset, and types using video and EEG recordings per the International League Against Epilepsy 2017 classification. We applied three neural network models-a convolutional neural network (CNN) and a CNN-long short-term memory (LSTM)-based generalized detection model and an autoencoder-based personalized detection model-to the raw time-series sensor data to detect seizures and utilized performance measures, including sensitivity, false positive rate (the number of false alarms divided by the total number of nonseizure segments), number of false alarms per day, and detection delay. We applied a 10-fold patientwise cross-validation scheme to the multisignal biosensor data and evaluated model performance on 28 seizure types. RESULTS: We analyzed 166 patients (47.6% female, median age = 10.0 years) and 900 seizures (13 254 h of sensor data) for 28 seizure types. With a CNN-LSTM-based seizure detection model, ACC, BVP, and their fusion performed better than chance; ACC and BVP data fusion reached the best detection performance of 83.9% sensitivity and 35.3% false positive rate. Nineteen of 28 seizure types could be detected by at least one data modality with area under receiver operating characteristic curve > .8 performance. SIGNIFICANCE: Results from this in-hospital study contribute to a paradigm shift in epilepsy care that entails noninvasive seizure detection, provides time-sensitive and accurate data on additional clinical seizure types, and proposes a novel combination of an out-of-the-box monitoring algorithm with an individualized person-oriented seizure detection approach.

Wearable device assessments of antiseizure medication effects on diurnal patterns of electrodermal activity, heart rate, and heart rate variability.

Patient-generated health data provide a great opportunity for more detailed ambulatory monitoring and more personalized treatments in many diseases. In epilepsy, robust diagnostics applicable to the ambulatory setting are needed as diagnosis and treatment decisions in current clinical practice are primarily reliant on patient self-reports, which are often inaccurate. Recent work using wearable devices has focused on methods to detect and forecast epileptic seizures. Whether wearable device signals may also contain information about the effect of antiseizure medications (ASMs), which may ultimately help to better monitor their efficacy, has not been evaluated yet. Here we systematically investigated the effect of ASMs on different data modalities (electrodermal activity, EDA, heart rate, HR, and heart rate variability, HRV) simultaneously recorded by a wearable device in 48 patients with epilepsy over several days in the epilepsy long-term monitoring unit at a tertiary hospital. All signals exhibited characteristic diurnal variations. HRV, but not HR or EDA-based metrics, were reduced by ASMs. By assessing multiple signals related to the autonomic nervous system simultaneously, our results provide novel insights into the effects of ASMs on the sympathetic and parasympathetic interplay in the setting of epilepsy and indicate the potential of easy-to-wear wearable devices for monitoring ASM action. Future work using longer data may investigate these metrics on multidien cycles and their utility for detecting seizures, assessing seizure risk, or informing treatment interventions.

Seizure detection using wearable sensors and machine learning: Setting a benchmark.

OBJECTIVE: Tracking seizures is crucial for epilepsy monitoring and treatment evaluation. Current epilepsy care relies on caretaker seizure diaries, but clinical seizure monitoring may miss seizures. Wearable devices may be better tolerated and more suitable for long-term ambulatory monitoring. This study evaluates the seizure detection performance of custom-developed machine learning (ML) algorithms across a broad spectrum of epileptic seizures utilizing wrist- and ankle-worn multisignal biosensors. METHODS: We enrolled patients admitted to the epilepsy monitoring unit and asked them to wear a wearable sensor on either their wrists or ankles. The sensor recorded body temperature, electrodermal activity, accelerometry (ACC), and photoplethysmography, which provides blood volume pulse (BVP). We used electroencephalographic seizure onset and offset as determined by a board-certified epileptologist as a standard comparison. We trained and validated ML for two different algorithms: Algorithm 1, ML methods for developing seizure type-specific detection models for nine individual seizure types; and Algorithm 2, ML methods for building general seizure type-agnostic detection, lumping together all seizure types. RESULTS: We included 94 patients (57.4% female, median age = 9.9 years) and 548 epileptic seizures (11 066 h of sensor data) for a total of 930 seizures and nine seizure types. Algorithm 1 detected eight of nine seizure types better than chance (area under the receiver operating characteristic curve [AUC-ROC] = .648-.976). Algorithm 2 detected all nine seizure types better than chance (AUC-ROC = .642-.995); a fusion of ACC and BVP modalities achieved the best AUC-ROC (.752) when combining all seizure types together. SIGNIFICANCE: Automatic seizure detection using ML from multimodal wearable sensor data is feasible across a broad spectrum of epileptic seizures. Preliminary results show better than chance seizure detection. The next steps include validation of our results in larger datasets, evaluation of the detection utility tool for additional clinical seizure types, and integration of additional clinical information.

CSNK2B: A broad spectrum of neurodevelopmental disability and epilepsy severity.

CSNK2B has recently been implicated as a disease gene for neurodevelopmental disability (NDD) and epilepsy. Information about developmental outcomes has been limited by the young age and short follow-up for many of the previously reported cases, and further delineation of the spectrum of associated phenotypes is needed. We present 25 new patients with variants in CSNK2B and refine the associated NDD and epilepsy phenotypes. CSNK2B variants were identified by research or clinical exome sequencing, and investigators from different centers were connected via GeneMatcher. Most individuals had developmental delay and generalized epilepsy with onset in the first 2 years. However, we found a broad spectrum of phenotypic severity, ranging from early normal development with pharmacoresponsive seizures to profound intellectual disability with intractable epilepsy and recurrent refractory status epilepticus. These findings suggest that CSNK2B should be considered in the diagnostic evaluation of patients with a broad range of NDD with treatable or intractable seizures.

Factors associated with long-term outcomes in pediatric refractory status epilepticus.

OBJECTIVE: This study was undertaken to describe long-term clinical and developmental outcomes in pediatric refractory status epilepticus (RSE) and identify factors associated with new neurological deficits after RSE. METHODS: We performed retrospective analyses of prospectively collected observational data from June 2011 to March 2020 on pediatric patients with RSE. We analyzed clinical outcomes from at least 30 days after RSE and, in a subanalysis, we assessed developmental outcomes and evaluated risk factors in previously normally developed patients. RESULTS: Follow-up data on outcomes were available in 276 patients (56.5% males). The median (interquartile range [IQR]) follow-up duration was 1.6 (.9-2.7) years. The in-hospital mortality rate was 4% (16/403 patients), and 15 (5.4%) patients had died after hospital discharge. One hundred sixty-six (62.9%) patients had subsequent unprovoked seizures, and 44 (16.9%) patients had a repeated RSE episode. Among 116 patients with normal development before RSE, 42 of 107 (39.3%) patients with available data had new neurological deficits (cognitive, behavioral, or motor). Patients with new deficits had longer median (IQR) electroclinical RSE duration than patients without new deficits (10.3 [2.1-134.5] h vs. 4 [1.6-16] h, p = .011, adjusted odds ratio = 1.003, 95% confidence interval = 1.0008-1.0069, p = .027). The proportion of patients with an unfavorable functional outcome (Glasgow Outcome Scale-Extended score ≥ 4) was 22 of 90 (24.4%), and they were more likely to have received a continuous infusion. SIGNIFICANCE: About one third of patients without prior epilepsy developed recurrent unprovoked seizures after the RSE episode. In previously normally developing patients, 39% presented with new deficits during follow-up, with longer electroclinical RSE duration as a predictor.

Benzodiazepine administration patterns before escalation to second-line medications in pediatric refractory convulsive status epilepticus.

OBJECTIVE: This study was undertaken to evaluate benzodiazepine (BZD) administration patterns before transitioning to non-BZD antiseizure medication (ASM) in pediatric patients with refractory convulsive status epilepticus (rSE). METHODS: This retrospective multicenter study in the United States and Canada used prospectively collected observational data from children admitted with rSE between 2011 and 2020. Outcome variables were the number of BZDs given before the first non-BZD ASM, and the number of BZDs administered after 30 and 45 min from seizure onset and before escalating to non-BZD ASM. RESULTS: We included 293 patients with a median (interquartile range) age of 3.8 (1.3-9.3) years. Thirty-six percent received more than two BZDs before escalating, and the later the treatment initiation was after seizure onset, the less likely patients were to receive multiple BZD doses before transitioning (incidence rate ratio [IRR] = .998, 95% confidence interval [CI] = .997-.999 per minute, p = .01). Patients received BZDs beyond 30 and 45 min in 57.3% and 44.0% of cases, respectively. Patients with out-of-hospital seizure onset were more likely to receive more doses of BZDs beyond 30 min (IRR = 2.43, 95% CI = 1.73-3.46, p < .0001) and beyond 45 min (IRR = 3.75, 95% CI = 2.40-6.03, p < .0001) compared to patients with in-hospital seizure onset. Intermittent SE was a risk factor for more BZDs administered beyond 45 min compared to continuous SE (IRR = 1.44, 95% CI = 1.01-2.06, p = .04). Forty-seven percent of patients (n = 94) with out-of-hospital onset did not receive treatment before hospital arrival. Among patients with out-of-hospital onset who received at least two BZDs before hospital arrival (n = 54), 48.1% received additional BZDs at hospital arrival. SIGNIFICANCE: Failure to escalate from BZDs to non-BZD ASMs occurs mainly in out-of-hospital rSE onset. Delays in the implementation of medical guidelines may be reduced by initiating treatment before hospital arrival and facilitating a transition to non-BZD ASMs after two BZD doses during handoffs between prehospital and in-hospital settings.

Association between semiologic, autonomic, and electrographic seizure characteristics in children with generalized tonic-clonic seizures.

INTRODUCTION: Generalized tonic-clonic seizures (GTCS) are associated with elevated electrodermal activity (EDA) and postictal generalized electroencephalographic suppression (PGES), markers that may indicate sudden unexpected death in epilepsy (SUDEP) risk. This study investigated the association of GTCS semiology, EDA, and PGES in children with epilepsy. METHODS: Patients admitted to the Boston Children's Hospital long-term video-EEG monitoring unit wore a sensor that records EDA. We selected patients with at least one GTCS and reviewed video-EEGs for semiology, tonic and clonic phase duration, total clinical seizure duration, electrographic onset, offset, and PGES. We grouped patients into three semiology classes: GTCS 1: bilateral symmetric tonic arm extension, GTCS 2: no specific tonic arm extension or flexion, GTCS 3: unilateral or asymmetrical arm extension, tonic arm flexion or posturing that does not fit into GTCS 1 or 2. We analyzed the correlation between semiology, EDA, and PGES, and measured the area under the curve (AUC) of the ictal EDA (seizure onset to one hour after), subtracting baseline EDA (one-hour seizure-free before seizure onset). Using generalized estimating equation (GEE) and linear regression, we analyzed all seizures and single episodes per patient. RESULTS: We included 30 patients (median age 13.8 ±â€¯3.6 years, 46.7% females) and 53 seizures. With GEE, GTCS 1 was associated with longer PGES duration compared to GTCS 2 (Estimate (ß) = -26.32 s, 95% Confidence Interval (CI): -36.46 to -16.18, p < 0.001), and the presence of PGES was associated with greater EDA change (ß = 429604 µS, 95% CI: 3550.96 to 855657.04, p = 0.048). With single-episode analysis, GTCS 1 had greater EDA change than GTCS 2 ((ß = -601339 µS, 95% CI: -1167016.56 to -35661.44, p = 0.047). EDA increased with PGES presence (ß = 637500 µS, 95% CI: 183571.84 to 1091428.16, p = 0.01) and duration (ß = 16794 µS, 95% CI: 5729.8 to 27858.2, p = 0.006). Patients with GTCS 1 had longer PGES duration compared to GTCS 2 (ß = -30.53 s, 95% CI: -44.6 to -16.46, p < 0.001) and GTCS 3 (ß = -22.07 s, 95% CI: -38.95 to -5.19, p = 0.016). CONCLUSION: In children with epilepsy, PGES correlates with greater ictal EDA. GTCS 1 correlated with longer PGES duration and may indirectly correlate with greater ictal EDA. Our study suggests potential applications in monitoring and preventing SUDEP in these patients.

Creation of a Novel Child Simulator and Curriculum to Optimize Administration of Seizure Rescue Medication.

INTRODUCTION: Although most cases of pediatric convulsive status epilepticus start in the prehospital setting, many patients do not receive treatment. The use of prehospital seizure rescue medications by caregivers is crucial, but studies suggest a lack of proper training on medication use. METHODS: We created a novel proof of principle mannequin and simulation for training proper administration of rectal diazepam, with a scoring paradigm to standardize and assess the educational process. RESULTS: Twenty-three health care providers (nurses and nurse practitioners, residents/fellows, and attending physicians) and 5 patient guardians/parents were included in the study. The rectal diazepam simulator displayed a high degree of physical and emotional realism (mean ≥ 4/5 on Likert scale survey) that effectively decreased time to treatment (-12.3 seconds; SD, 16.3) and improved the accuracy of medication delivery in a simulation setting (-4.2 points; SD, 3.1). The scoring technique had appropriate interrater reliability (≥86% on all but 2 prompts) and was a feasible instrument to assess the effectiveness of the educational intervention. CONCLUSIONS: A unique procedure-focused child simulator and rescue medication score offer an innovative and effective means to train caregivers on the use of lifesaving seizure rescue medications.

Pediatric status epilepticus management by Emergency Medical Services (the pSERG cohort).

PURPOSE: Delayed treatment in status epilepticus (SE) is independently associated with increased treatment resistance, morbidity, and mortality. We describe the prehospital management pathway and Emergency Medical Services (EMS) timeliness in children who developed refractory convulsive status epilepticus (RCSE). METHODS: Retrospective multicenter study in the United States using prospectively collected observational data from June 2011 to March 2020. We selected pediatric patients (one month-21 years) with RCSE initiated outside the hospital and transported to the hospital by EMS. RESULTS: We included 91 patients with a median (percentile25-percentile75) age of 3.0 (1.5-7.3) years. The median time from seizure onset to hospital arrival was 45 (30-67) minutes, with a median time cared for by EMS of 24 (15-36) minutes. Considering treatment by caregivers and EMS before hospital arrival, 20 (22%) patients did not receive any anti-seizure medications (ASM) and 71 (78%) received one to five doses of benzodiazepines (BZD), without non-BZD ASM. We provided the prehospital treatment flow path of these patients through caregivers and EMS including relevant time points. Patients with a history of SE were more likely to receive the first BZD in the prehospital setting compared to patients without a history of SE (adjusted HR 3.25, 95% CI 1.72-6.12, p<0.001). CONCLUSION: In this multicenter study of pediatric RCSE, prehospital treatment may be streamlined further. Patients with a history of SE were more likely to receive prehospital rescue medication.

Seizure-related differences in biosignal 24-h modulation patterns.

A seizure likelihood biomarker could improve seizure monitoring and facilitate adjustment of treatments based on seizure risk. Here, we tested differences in patient-specific 24-h-modulation patterns of electrodermal activity (EDA), peripheral body temperature (TEMP), and heart rate (HR) between patients with and without seizures. We enrolled patients who underwent continuous video-EEG monitoring at Boston Children's Hospital to wear a biosensor. We divided patients into two groups: those with no seizures and those with at least one seizure during the recording period. We assessed the 24-h modulation level and amplitude of EDA, TEMP, and HR. We performed machine learning including physiological and clinical variables. Subsequently, we determined classifier performance by cross-validated machine learning. Patients with seizures (n = 49) had lower EDA levels (p = 0.031), EDA amplitudes (p = 0.045), and trended toward lower HR levels (p = 0.060) compared to patients without seizures (n = 68). Averaged cross-validated classification accuracy was 69% (AUC-ROC: 0.75). Our results show the potential to monitor and forecast risk for epileptic seizures based on changes in 24-h patterns in wearable recordings in combination with clinical variables. Such biomarkers might be applicable to inform care, such as treatment or seizure injury risk during specific periods, scheduling diagnostic tests, such as admission to the epilepsy monitoring unit, and potentially other neurological and chronic conditions.

Time to Treatment in Pediatric Convulsive Refractory Status Epilepticus: The Weekend Effect.

BACKGROUND: Time to treatment in pediatric refractory status epilepticus is delayed. We aimed to evaluate the influence of weekends and holidays on time to treatment of this pediatric emergency. METHODS: We performed a retrospective analysis of prospectively collected observational data of pediatric patients with refractory status epilepticus. RESULTS: We included 329 patients (56% males) with a median (p25 to p75) age of 3.8 (1.3 to 9) years. The median (p25 to p75) time to first BZD on weekdays and weekends/holidays was 20 (6.8 to 48.3) minutes versus 11 (5 to 35) minutes, P = 0.01; adjusted hazard ratio (HR) = 1.20 (95% confidence interval [CI]: 0.95 to 1.55), P = 0.12. The time to first non-BZD ASM was longer on weekdays than on weekends/holidays (68 [42.8 to 153.5] minutes versus 59 [27 to 120] minutes, P = 0.006; adjusted HR = 1.38 [95% CI: 1.08 to 1.76], P = 0.009). However, this difference was mainly driven by status epilepticus with in-hospital onset: among 108 patients, the time to first non-BZD ASM was longer during weekdays than during weekends/holidays (55.5 [28.8 to 103.5] minutes versus 28 [15.8 to 66.3] minutes, P = 0.003; adjusted HR = 1.65 [95% CI: 1.08 to 2.51], P = 0.01). CONCLUSIONS: The time to first non-BZD ASM in pediatric refractory status epilepticus is shorter on weekends/holidays than on weekdays, mainly driven by in-hospital onset status epilepticus. Data on what might be causing this difference may help tailor policies to improve medication application timing.