Search

OBJECTIVE: Video-electroencephalography (vEEG) is an important component of epilepsy diagnosis and management. Nevertheless, inpatient vEEG monitoring fails to capture seizures in up to one third of patients. We hypothesized that personalized seizure forecasts could be used to optimize the timing of vEEG. METHODS: We used a database of ambulatory vEEG studies to select a cohort with linked electronic seizure diaries of more than 20 reported seizures over at least 8 weeks. The total cohort included 48 participants. Diary seizure times were used to detect individuals' multiday seizure cycles and estimate times of high seizure risk. We compared whether estimated seizure risk was significantly different between conclusive and inconclusive vEEGs, and between vEEG with and without recorded epileptic activity. vEEGs were conducted prior to self-reported seizures; hence, the study aimed to provide a retrospective proof of concept that cycles of seizure risk were correlated with vEEG outcomes. RESULTS: Estimated seizure risk was significantly higher for conclusive vEEGs and vEEGs with epileptic activity. Across all cycle strengths, the average time in high risk during vEEG was 29.1% compared with 14% for the conclusive/inconclusive groups and 32% compared to 18% for the epileptic activity/no epileptic activity groups. On average, 62.5% of the cohort showed increased time in high risk during their previous vEEG when epileptic activity was recorded (compared to 28% of the cohort where epileptic activity was not recorded). For conclusive vEEGs, 50% of the cohort had increased time in high risk, compared to 21.5% for inconclusive vEEGs. SIGNIFICANCE: Although retrospective, this study provides a proof of principle that scheduling monitoring times based on personalized seizure risk forecasts can improve the yield of vEEG. Forecasts can be developed at low cost from mobile seizure diaries. A simple scheduling tool to improve diagnostic outcomes may reduce cost and risks associated with delayed or missed diagnosis in epilepsy.

© 2020 The Authors. Annals of Clinical and Translational Neurology published by Wiley Periodicals LLC on behalf of American Neurological Association. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited., We describe the longest period of subcutaneous EEG (sqEEG) monitoring to date, in a 35-year-old female with refractory epilepsy. Over 230 days, 4791/5520 h of sqEEG were recorded (86%, mean 20.8 [IQR 3.9] hours/day). Using an electronic diary, the patient reported 22 seizures, while automatically-assisted visual sqEEG review detected 32 seizures. There was substantial agreement between days of reported and recorded seizures (Cohen's kappa 0.664), although multiple clustered seizures remained undocumented. Circular statistics identified significant sqEEG seizure cycles at circadian (24-hour) and multidien (5-day) timescales. Electrographic seizure monitoring and analysis of long-term seizure cycles are possible with this neurophysiological tool., This work was supported by the Epilepsy Foundation’s Epilepsy Innovation Institute My Seizure Gauge Project. MPR is supported by the NIHR Biomedical Research Centre; the MRC Centre for Neurodevelopmental Disorders (MR/N026063/1); the EPSRC Centre for Predictive Modelling in Healthcare (EP/N014391/1); the RADAR‐CNS project (www.radar‐cns.org, grant agreement 115902).

Objective Seizure unpredictability is rated as one of the most challenging aspects of living with epilepsy. Seizure likelihood can be influenced by a range of environmental and physiological factors that are difficult to measure and quantify. However, some generalizable patterns have been demonstrated in seizure onset. A majority of people with epilepsy exhibit circadian rhythms in their seizure times, and many also show slower, multiday patterns. Seizure cycles can be measured using a range of recording modalities, including self-reported electronic seizure diaries. This study aimed to develop personalized forecasts from a mobile seizure diary app. Methods Forecasts based on circadian and multiday seizure cycles were tested pseudoprospectively using data from 50 app users (mean of 109 seizures per subject). Individuals' strongest cycles were estimated from their reported seizure times and used to derive the likelihood of future seizures. The forecasting approach was validated using self-reported events and electrographic seizures from the Neurovista dataset, an existing database of long-term electroencephalography that has been widely used to develop forecasting algorithms. Results The validation dataset showed that forecasts of seizure likelihood based on self-reported cycles were predictive of electrographic seizures for approximately half the cohort. Forecasts using only mobile app diaries allowed users to spend an average of 67.1% of their time in a low-risk state, with 14.8% of their time in a high-risk warning state. On average, 69.1% of seizures occurred during high-risk states and 10.5% of seizures occurred in low-risk states. Significance Seizure diary apps can provide personalized forecasts of seizure likelihood that are accurate and clinically relevant for electrographic seizures. These results have immediate potential for translation to a prospective seizure forecasting trial using a mobile diary app. It is our hope that seizure forecasting apps will one day give people with epilepsy greater confidence in managing their daily activities.

The ability to forecast seizures minutes to hours in advance of an event has been verified using invasive EEG devices, but has not been previously demonstrated using noninvasive wearable devices over long durations in an ambulatory setting. In this study we developed a seizure forecasting system with a long short-term memory (LSTM) recurrent neural network (RNN) algorithm, using a noninvasive wrist-worn research-grade physiological sensor device, and tested the system in patients with epilepsy in the field, with concurrent invasive EEG confirmation of seizures via an implanted recording device. The system achieved forecasting performance significantly better than a random predictor for 5 of 6 patients studied, with mean AUC-ROC of 0.80 (range 0.72–0.92). These results provide the first clear evidence that direct seizure forecasts are possible using wearable devices in the ambulatory setting for many patients with epilepsy.

BACKGROUND: Circadian and multiday rhythms are found across many biological systems, including cardiology, endocrinology, neurology, and immunology. In people with epilepsy, epileptic brain activity and seizure occurrence have been found to follow circadian, weekly, and monthly rhythms. Understanding the relationship between these cycles of brain excitability and other physiological systems can provide new insight into the causes of multiday cycles. The brain-heart link has previously been considered in epilepsy research, with potential implications for seizure forecasting, therapy, and mortality (i.e., sudden unexpected death in epilepsy). METHODS: We report the results from a non-interventional, observational cohort study, Tracking Seizure Cycles. This study sought to examine multiday cycles of heart rate and seizures in adults with diagnosed uncontrolled epilepsy (N=31) and healthy adult controls (N=15) using wearable smartwatches and mobile seizure diaries over at least four months (M=12.0, SD=5.9; control M=10.6, SD=6.4). Cycles in heart rate were detected using a continuous wavelet transform. Relationships between heart rate cycles and seizure occurrence were measured from the distributions of seizure likelihood with respect to underlying cycle phase. FINDINGS: Heart rate cycles were found in all 46 participants (people with epilepsy and healthy controls), with circadian (N=46), about-weekly (N=25) and about-monthly (N=13) rhythms being the most prevalent. Of the participants with epilepsy, 19 people had at least 20 reported seizures, and 10 of these had seizures significantly phase locked to their multiday heart rate cycles. INTERPRETATION: Heart rate cycles showed similarities to multiday epileptic rhythms and may be comodulated with seizure likelihood. The relationship between heart rate and seizures is relevant for epilepsy therapy, including seizure forecasting, and may also have implications for cardiovascular disease. More broadly, understanding the link between multiday cycles in the heart and brain can shed new light on endogenous physiological rhythms in humans. FUNDING: This research received funding from the Australian Government National Health and Medical Research Council (investigator grant 1178220), the Australian Government BioMedTech Horizons program, and the Epilepsy Foundation of America's 'My Seizure Gauge' grant. ispartof: EBIOMEDICINE vol:72 ispartof: location:Netherlands status: published

Accurate identification of seizure activity, both clinical and subclinical, has important implications in the management of epilepsy. Accurate recognition of seizure activity is essential for diagnostic, management and forecasting purposes, but patient-reported seizures have been shown to be unreliable. Earlier work has revealed accurate capture of electrographic seizures and forecasting is possible with an implantable intracranial device, but less invasive electroencephalography (EEG) recording systems would be optimal. Here, we present preliminary results of seizure detection and forecasting with a minimally invasive sub-scalp device that continuously records EEG. Five participants with refractory epilepsy who experience at least two clinically identifiable seizures monthly have been implanted with sub-scalp devices (Minder®), providing two channels of data from both hemispheres of the brain. Data is continuously captured via a behind-the-ear system, which also powers the device, and transferred wirelessly to a mobile phone, from where it is accessible remotely via cloud storage. EEG recordings from the sub-scalp device were compared to data recorded from a conventional system during a 1-week ambulatory video-EEG monitoring session. Suspect epileptiform activity (EA) was detected using machine learning algorithms and reviewed by trained neurophysiologists. Seizure forecasting was demonstrated retrospectively by utilizing cycles in EA and previous seizure times. The procedures and devices were well-tolerated and no significant complications have been reported. Seizures were accurately identified on the sub-scalp system, as visually confirmed by periods of concurrent conventional scalp EEG recordings. The data acquired also allowed seizure forecasting to be successfully undertaken. The area under the receiver operating characteristic curve (AUC score) achieved (0.88), which is comparable to the best score in recent, state-of-the-art forecasting work using intracranial EEG.

SummaryObjectiveUltra long-term subcutaneous EEG (sqEEG) monitoring is a new modality with great potential for both health and disease, including epileptic seizure detection and forecasting. However, little is known about the long-term quality and consistency of the sqEEG signal, which is the objective of this study.MethodsThe largest multicenter cohort of sqEEG was analyzed, including fourteen patients with epilepsy and twelve healthy subjects, implanted with a sqEEG device (24/7 EEG™ SubQ), and recorded from 23 to 230 days (median 42 days), with a median data capture rate of 75% (17.9 hours/day). Median power spectral density plots of each subject were examined for physiological peaks, including at diurnal and nocturnal periods. Long-term temporal trends in signal impedance and power spectral features were investigated with subject-specific linear regression models and group-level linear mixed effects models.ResultssqEEG spectrograms showed an approximately 1/f power distribution. Diurnal peaks in the alpha range (8-13Hz) and nocturnal peaks in the sigma range (12-16Hz) were seen in the majority of subjects. Signal impedances remained low and frequency band powers were highly stable throughout the recording periods.SignificanceThe spectral characteristics of minimally-invasive, ultra long-term sqEEG are similar to scalp EEG, while the signal is highly stationary. Our findings reinforce the suitability of this system for chronic implantation on diverse clinical applications, from seizure detection and forecasting to brain-computer interfaces.Key PointsSubcutaneous EEG shows similar spectral characteristics to scalp EEGThe subcutaneous EEG signal is highly stable throughout weeks and months of recordingSubcutaneous EEG systems are well suited for chronic implantation, for seizure detection and seizure forecasting

© Copyright 2021 IOP Publishing., Objective. The detection of seizures using wearable devices would improve epilepsy management, but reliable detection of seizures in an ambulatory environment remains challenging, and current studies lack concurrent validation of seizures using electroencephalography (EEG) data. Approach. An adaptively trained long–short-term memory deep neural network was developed and trained using a modest number of seizure data sets from wrist-worn devices. Transfer learning was used to adapt a classifier that was initially trained on intracranial electroencephalography (iEEG) signals to facilitate classification of non-EEG physiological datasets comprising accelerometry, blood volume pulse, skin electrodermal activity, heart rate, and temperature signals. The algorithm’s performance was assessed with and without pre-training on iEEG signals and transfer learning. To assess the performance of the seizure detection classifier using long-term ambulatory data, wearable devices were used for multiple months with an implanted neurostimulator capable of recording iEEG signals, which provided independent electrographic seizure detections that were reviewed by a board-certified epileptologist. Main results. For 19 motor seizures from 10 in-hospital patients, the algorithm yielded a mean area under curve (AUC), a sensitivity, and an false alarm rate per day (FAR/day) of 0.98, 0.93, and 2.3, respectively. Additionally, for eight seizures with probable motor semiology from two ambulatory patients, the classifier achieved a mean AUC of 0.97 and an FAR of 2.45 events/day at a sensitivity of 0.9. For all seizure types in the ambulatory setting, the classifier had a mean AUC of 0.82 with a sensitivity of 0.47 and an FAR of 7.2 events/day. Significance. The performance of the algorithm was evaluated using motor and non-motor seizures during in-hospital and ambulatory use. The classifier was able to detect multiple types of motor and non-motor seizures, but performed significantly better on motor seizures.

Copyright © 2021 Brinkmann, Karoly, Nurse, Dumanis, Nasseri, Viana, Schulze-Bonhage, Freestone, Worrell, Richardson and Cook. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms., It is a major challenge in clinical epilepsy to diagnose and treat a disease characterized by infrequent seizures based on patient or caregiver reports and limited duration clinical testing. The poor reliability of self-reported seizure diaries for many people with epilepsy is well-established, but these records remain necessary in clinical care and therapeutic studies. A number of wearable devices have emerged, which may be capable of detecting seizures, recording seizure data, and alerting caregivers. Developments in non-invasive wearable sensors to measure accelerometry, photoplethysmography (PPG), electrodermal activity (EDA), electromyography (EMG), and other signals outside of the traditional clinical environment may be able to identify seizure-related changes. Non-invasive scalp electroencephalography (EEG) and minimally invasive subscalp EEG may allow direct measurement of seizure activity. However, significant network and computational infrastructure is needed for continuous, secure transmission of data. The large volume of data acquired by these devices necessitates computer-assisted review and detection to reduce the burden on human reviewers. Furthermore, user acceptability of such devices must be a paramount consideration to ensure adherence with long-term device use. Such devices can identify tonic-clonic seizures, but identification of other seizure semiologies with non-EEG wearables is an ongoing challenge. Identification of electrographic seizures with subscalp EEG systems has recently been demonstrated over long (>6 month) durations, and this shows promise for accurate, objective seizure records. While the ability to detect and forecast seizures from ambulatory intracranial EEG is established, invasive devices may not be acceptable for many individuals with epilepsy. Recent studies show promising results for probabilistic forecasts of seizure risk from long-term wearable devices and electronic diaries of self-reported seizures. There may also be predictive value in individuals' symptoms, mood, and cognitive performance. However, seizure forecasting requires perpetual use of a device for monitoring, increasing the importance of the system's acceptability to users. Furthermore, long-term studies with concurrent EEG confirmation are lacking currently. This review describes the current evidence and challenges in the use of minimally and non-invasive devices for long-term epilepsy monitoring, the essential components in remote monitoring systems, and explores the feasibility to detect and forecast impending seizures via long-term use of these systems.

©Andrea Biondi, Petroula Laiou, Elisa Bruno, Pedro F Viana, Martijn Schreuder, William Hart, Ewan Nurse, Deb K Pal, Mark P Richardson. Originally published in JMIR Research Protocols (http://www.researchprotocols.org), 19.03.2021. This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Research Protocols, is properly cited. The complete bibliographic information, a link to the original publication on http://www.researchprotocols.org, as well as this copyright and license information must be included., Background: Epileptic seizures are spontaneous events that severely affect the lives of patients due to their recurrence and unpredictability. The integration of new wearable and mobile technologies to collect electroencephalographic (EEG) and extracerebral signals in a portable system might be the solution to prospectively identify times of seizure occurrence or propensity. The performances of several seizure detection devices have been assessed by validated studies, and patient perspectives on wearables have been explored to better match their needs. Despite this, there is a major gap in the literature on long-term, real-life acceptability and performance of mobile technology essential to managing chronic disorders such as epilepsy. Objective: EEG@HOME is an observational, nonrandomized, noninterventional study that aims to develop a new feasible procedure that allows people with epilepsy to independently, continuously, and safely acquire noninvasive variables at home. The data collected will be analyzed to develop a general model to predict periods of increased seizure risk. Methods: A total of 12 adults with a diagnosis of pharmaco-resistant epilepsy and at least 20 seizures per year will be recruited at King's College Hospital, London. Participants will be asked to self-apply an easy and portable EEG recording system (ANT Neuro) to record scalp EEG at home twice daily. From each serial EEG recording, brain network ictogenicity (BNI), a new biomarker of the propensity of the brain to develop seizures, will be extracted. A noninvasive wrist-worn device (Fitbit Charge 3; Fitbit Inc) will be used to collect non-EEG biosignals (heart rate, sleep quality index, and steps), and a smartphone app (Seer app; Seer Medical) will be used to collect data related to seizure occurrence, medication taken, sleep quality, stress, and mood. All data will be collected continuously for 6 months. Standardized questionnaires (the Post-Study System Usability Questionnaire and System Usability Scale) will be completed to assess the acceptability and feasibility of the procedure. BNI, continuous wrist-worn sensor biosignals, and electronic survey data will be correlated with seizure occurrence as reported in the diary to investigate their potential values as biomarkers of seizure risk. Results: The EEG@HOME project received funding from Epilepsy Research UK in 2018 and was approved by the Bromley Research Ethics Committee in March 2020. The first participants were enrolled in October 2020, and we expect to publish the first results by the end of 2022. Conclusions: With the EEG@HOME study, we aim to take advantage of new advances in remote monitoring technology, including self-applied EEG, to investigate the feasibility of long-term disease self-monitoring. Further, we hope our study will bring new insights into noninvasively collected personalized risk factors of seizure occurrence and seizure propensity that may help to mitigate one of the most difficult aspects of refractory epilepsy: the unpredictability of seizure occurrence, This study is funded by Epilepsy Research UK (award 1803). MPR, PFV, and EN are supported by the Epilepsy Foundation of America’s Epilepsy Innovation Institute My Seizure Gauge grant.

ObjectiveVideo-electroencephalography (vEEG) is an important component of epilepsy diagnosis and management. Nevertheless, inpatient vEEG monitoring fails to capture seizures in up to one third of patients during diagnostic and pre-surgical monitoring. We hypothesized that personalized seizure forecasts could be used to optimize the timing of vEEG and improve diagnostic yield. MethodsWe used a database of ambulatory vEEG studies to select a cohort with linked electronic seizure diaries of more than 20 reported seizures over at least 8 weeks. The total cohort included 48 participants. Diary seizure times were used to detect individuals multi-day seizure cycles and estimate times of high seizure risk. We then compared whether estimated seizure risk was significantly different between diagnostic and non-diagnostic vEEGs, and between vEEG with and without recorded epileptic activity. ResultsEstimated seizure risk was significantly higher for diagnostic vEEGs and vEEGs with epileptic activity. Across all cycle strengths, the average time in high risk during vEEG was 29.1% compared with 14% for the diagnostic/non-diagnostic groups and 32% compared to 18% for the epileptic activity/no epileptic activity groups. On average, 62.5% of the cohort showed increased time in high risk during vEEG when epileptic activity was recorded (compared to 28% of the cohort where epileptic activity was not recorded). For diagnostic vEEGs, 50% of the cohort had increased time in high risk, compared to 21.5% for non-diagnostic vEEGs. SignificanceThis study provides a proof of principle that scheduling monitoring times based on personalized seizure risk forecasts can improve the yield of vEEG. Importantly, forecasts can be developed at low cost from mobile seizure diaries. A simple scheduling tool to improve diagnostic outcomes has the potential to reduce the significant cost and risks associated with delayed or missed diagnosis in epilepsy. Key PointsO_LISelf-reported seizure cycles, measured from electronic diaries, correspond to outcomes of diagnostic video-EEG monitoring C_LIO_LICycle-based forecasts of seizure risk were related to both diagnostic yield and occurrence of epileptic activity during monitoring C_LIO_LIPersonalized forecasts can be used to schedule video-EEG monitoring to coincide with periods of heighted seizure risk C_LIO_LISmart scheduling of video-EEG monitoring could improve yield and reduce the burden of delayed or missed diagnoses in epilepsy C_LI

Machine learning (ML) is increasingly recognized as a useful tool in healthcare applications, including epilepsy. One of the most important applications of ML in epilepsy is seizure detection and prediction, using wearable devices (WDs). However, not all currently available algorithms implemented in WDs are using ML. In this review, we summarize the state of the art of using WDs and ML in epilepsy, and we outline future development in these domains. There is published evidence for reliable detection of epileptic seizures using implanted electroencephalography (EEG) electrodes and wearable, non-EEG devices. Application of ML using the data recorded with WDs from a large number of patients could change radically the way we diagnose and manage patients with epilepsy.

Objective We report on patient-specific durations of postictal periods in long-term intracranial electroencephalography (iEEG) recordings. The objective was to investigate the relationship between seizure duration and postictal suppression duration. Methods Long-term recording iEEG from 9 patients (>50 seizures recorded) were analyzed. In total, 2310 seizures were recorded during a total of 13.8 years of recording. Postictal suppression duration was calculated as the duration after seizure termination until total signal energy returned to background levels. The relationship between seizure duration and postictal suppression duration was quantified using the correlation coefficient (r). The effects of populations of seizures within patients, on correlations, were also considered. Populations of seizures within patients were distinguished by seizure duration thresholds and k-means clustering along the dimensions of seizure duration and postictal suppression duration. The effects of bursts of seizures were also considered by defining populations based on interseizure interval (ISI). Results Seizure duration accounted for 40% of postictal suppression duration variance, aggregated across all patients and seizures. Seizure duration accounted for more than 25% of the variance in postictal suppression duration in 2 patients and accounted for less than 25% in the remaining 7. In 3 patients, heat maps showed multiple distinct postictal patterns indicating multiple populations of seizures. When accounting for these populations, seizure duration accounted for less than 25% of the variance in postictal duration in all populations. Variance in postictal suppression duration accounted for less than 10% of ISI variance in all patients. Significance We have previously demonstrated that some patients have multiple seizure populations distinguishable by seizure duration. This article shows that different seizure populations have distinct and consistent postictal behaviors. The existence of multiple populations in some patients has implications for seizure management and forecasting, whereas the distinct postictal behaviors may have implications for sudden unexpected death in epilepsy (SUDEP) prediction and prevention.

Epilepsy diagnosis can be costly, time-consuming, and not uncommonly inaccurate. The reference standard diagnostic monitoring is continuous video-electroencephalography (EEG) monitoring, ideally capturing all events or concordant interictal discharges. Automating EEG data review would save time and resources, thus enabling more people to receive reference standard monitoring and also potentially heralding a more quantitative approach to therapeutic outcomes. There is substantial research into the automated detection of seizures and epileptic activity from EEG. However, automated detection software is not widely used in the clinic, and despite numerous published algorithms, few methods have regulatory approval for detecting epileptic activity from EEG. This study reports on a deep learning algorithm for computer-assisted EEG review. Deep convolutional neural networks were trained to detect epileptic discharges using a preexisting dataset of over 6000 labelled events in a cohort of 103 patients with idiopathic generalized epilepsy (IGE). Patients underwent 24-hour ambulatory outpatient EEG, and all data were curated and confirmed independently by two epilepsy specialists (Seneviratne et al., 2016). The resulting automated detection algorithm was then used to review diagnostic scalp EEG for seven patients (four with IGE and three with events mimicking seizures) to validate performance in a clinical setting. The automated detection algorithm showed state-of-the-art performance for detecting epileptic activity from clinical EEG, with mean sensitivity of >95% and corresponding mean false positive rate of 1 detection per minute. Importantly, diagnostic case studies showed that the automated detection algorithm reduced human review time by 80%-99%, without compromising event detection or diagnostic accuracy. The presented results demonstrate that computer-assisted review can increase the speed and accuracy of EEG assessment and has the potential to greatly improve therapeutic outcomes. This article is part of the Special Issue "NEWroscience 2018".

Background Seizure prediction can increase independence and allow preventative treatment for patients with epilepsy. We present a proof-of-concept for a seizure prediction system that is accurate, fully automated, patient-specific, and tunable to an individual's needs. Methods Intracranial electroencephalography (iEEG) data of ten patients obtained from a seizure advisory system were analyzed as part of a pseudoprospective seizure prediction study. First, a deep learning classifier was trained to distinguish between preictal and interictal signals. Second, classifier performance was tested on held-out iEEG data from all patients and benchmarked against the performance of a random predictor. Third, the prediction system was tuned so sensitivity or time in warning could be prioritized by the patient. Finally, a demonstration of the feasibility of deployment of the prediction system onto an ultra-low power neuromorphic chip for autonomous operation on a wearable device is provided. Results The prediction system achieved mean sensitivity of 69% and mean time in warning of 27%, significantly surpassing an equivalent random predictor for all patients by 42%. Conclusion This study demonstrates that deep learning in combination with neuromorphic hardware can provide the basis for a wearable, real-time, always-on, patient-specific seizure warning system with low power consumption and reliable long-term performance., Highlights • We use deep learning and long-term neural data to develop an automated, patient-tunable epileptic seizure prediction system. • We deploy our prediction system on a low-power neuromorphic chip to form the basis of a wearable device. Predicting and treating the debilitating seizures suffered by epileptic patients has challenged medical researchers for over fifty years. A new way forward was opened when Cook and colleagues, in 2013, collected a large longitudinal and continuous dataset recorded directly from patients' brains for one to three years. Harnessing the recent breakthroughs in deep learning techniques and in building specialized processing chips, we have demonstrated that seizures can now be predicted by a portable device. Our system automatically learns patient-specific pre-seizure signatures, and, in real time, warns of oncoming seizures.

Background: Circadian and multiday rhythms are found across many biological systems, including cardiology, endocrinology, neurology and immunology. In people with epilepsy, epileptic brain activity and seizure occurrence have been found to follow circadian, weekly and monthly rhythms. Understanding the relationship between these cycles of brain excitability and other physiological systems can provide new insight into the causes of multiday cycles. The brain-heart link is highly relevant for epilepsy, with implications for seizure prediction, therapy (i.e. vagal nerve stimulation) and mortality (i.e. sudden unexpected death in epilepsy). Methods: We report the results from a non-interventional, observational cohort study, Tracking Seizure Cycles. This study quantified multiday cycles of heart rate and seizures in human participants (N=26) using wearable smartwatches and mobile seizure diaries over at least two months (M=6.4, SD=4.0). Participants had an epilepsy diagnosis and uncontrolled or partially controlled seizures. Significant cycles in heart rate were detected using a continuous wavelet transform. Significant co-modulation between heart rate cycles and seizure occurrence were measured from the distributions of seizure likelihood with respect to underlying cycle phase (Omnibus test). Weekly trends were also investigated from a retrospective cohort of continuous video-EEG-ECG records of people (N=29) undergoing diagnostic epilepsy monitoring for at least eight days (M=9.7, SD=1.3). Findings: Heart rate cycles were found in all participants, with circadian (100%), about-weekly (88%) and about-monthly (92%) rhythms being the most prevalent. Seizures were significantly locked onto at least one heart rate cycle in 76% of eligible participants. Seizures occurred preferentially during the rising phase of the heart rate cycles. Significant weekly trends were also verified in 52% of the video-EEG-ECG cohort. Interpretation: The existence and prevalence of multiday heart rate cycles may have clinical implications for cardiology. Heart rate cycles showed striking similarities to multiday epileptic rhythms and were co-modulated with seizure likelihood. The relationship between heart rate and seizures is relevant for epilepsy therapy, including seizure forecasting. More broadly, understanding the link between multiday cycles in the heart and brain can shed new light on endogenous physiological rhythms in humans. Funding Statement: This research was funded by an NHMRC Investigator Grant 1178220. This research was supported by the Epilepsy Foundation of America's My Seizure Gauge grant. Declaration of Interests: Dr. Brinkmann reports grants from Epilepsy Foundation of America, My Seizure Gauge, during the conduct of the study; other from Cadence Neurosciences, outside the submitted work. Dr. Freestone reports grants from Epilepsy Foundation USA, personal fees and other from Seer Medical, during the conduct of the study; In addition, Dr. Freestone has a patent Methods and Systems of Seizure Forecasting issued. Dr. Nurse reports grants from Epilepsy Foundation of America, grants from MTPConnect, during the conduct of the study; personal fees from Seer Medical, outside the submitted work. Dr. Cook reports personal fees and other from Seer Medical Australia, personal fees and other from Epi Minder, outside the submitted work; In addition, Dr. Cook has a patent Methods and Systems of Seizure Forecasting issued. Dr. Karoly reports grants from National Health and Medical Research Council (NHMRC), during the conduct of the study; other from Seer Medical (Australia), outside the submitted work; In addition, Dr. Karoly has a patent Methods and Systems of Seizure Forecasting issued. The remaining authors have no competing interests to declare. Ethics Approval Statement: The study was approved by the St Vincent’s Hospital Human Research Ethics Committee (HREC 009.19) and first enrolment was in August 2019. All participants provided written informed consent.

SUMMARYObjectiveSeizure unpredictability is rated as one of the most challenging aspects of living with epilepsy. Seizure likelihood can be influenced by a range of environmental and physiological factors that are difficult to measure and quantify. However, some generalizable patterns have been demonstrated in seizure onset. A majority of people with epilepsy exhibit circadian rhythms in their seizure times and many also show slower, multiday patterns. Seizure cycles can be measured using a range of recording modalities, including self-reported electronic seizure diaries. This study aimed to develop personalized forecasts from a mobile seizure diary app.MethodsForecasts based on circadian and multiday seizure cycles were tested pseudo-prospectively using data from 33 app users (mean of 103 seizures per subject). Individual’s strongest cycles were estimated from their reported seizure times and used to derive the likelihood of future seizures. The forecasting approach was validated using self-reported events and electrographic seizures from the Neurovista dataset, an existing database of long-term electroencephalography that has been widely used to develop forecasting algorithms.ResultsThe validation dataset showed that forecasts of seizure likelihood based on self-reported cycles were predictive of electrographic seizures. Forecasts using only mobile app diaries allowed users to spend an average of 62.8% of their time in a low-risk state, with 16.6% of their time in a high-risk warning state. On average, 64.5% of seizures occurred during high-risk states and less than 10% of seizures occurred in low-risk states.SignificanceSeizure diary apps can provide personalized forecasts of seizure likelihood that are accurate and clinically relevant for electrographic seizures. These results have immediate potential for translation to a prospective seizure forecasting trial using a mobile diary app. It is our hope that seizure forecasting apps will one day give people with epilepsy greater confidence in managing their daily activities.

Electroencephalography (EEG) has been used to forecast seizures with varying success. There is an increasing interest to use electrocardiogram (ECG) to help with seizure forecasting. The neural and cardiovascular systems may exhibit critical slowing, which is measured by an increase in variance and autocorrelation of the system, when change from a normal state to an ictal state. To forecast seizures, the variance and autocorrelation of long-term continuous EEG and ECG data from 16 patients were used for analysis. The average period of recordings was 161.9 h, with an average of 9 electrographic seizures in an individual patient. The relationship between seizure onset times and phases of variance and autocorrelation in EEG and ECG data was investigated. The results of forecasting models using critical slowing features, seizure circadian features, and combined critical slowing and circadian features were compared using the receiver-operating characteristic curve. The results demonstrated that the best forecaster was patient-specific and the average area under the curve (AUC) of the best forecaster across patients was 0.68. In 50% of patients, circadian forecasters had the best performance. Critical slowing forecaster performed best in 19% of patients. Combined forecaster achieved the best performance in 31% of patients. The results of this study may help to advance the field of seizure forecasting and lead to the improved quality of life of people who suffer from epilepsy.

SummaryObjective We report on temporally clustered seizures detected from continuous long-term ambulatory human electroencephalographic data. The objective was to investigate short-term seizure clustering, which we have termed bursting, and consider implications for patient care, seizure prediction, and evaluating therapies. Methods Chronic ambulatory intracranial electroencephalography (EEG) data collected for the purpose of seizure prediction were annotated to identify seizure events. A detection algorithm was used to identify bursts of events. Burst events were compared to nonburst events to evaluate event dispersion, duration and dynamics. Results Bursts of seizures were present in 6 of 15 subjects, and detections were consistent over long-term monitoring (>2 years). Subjects with bursts of seizures had highly overdispersed seizure rates, compared to other subjects. There was a complicated relationship between bursts and clinical seizures, although bursts were associated with multimodal distributions of seizure duration, and poorer predictive outcomes. For three subjects, bursts demonstrated distinctive preictal dynamics compared to clinical seizures. Significance We have previously hypothesized that there are distinct physiologic pathways underlying short- and long-duration seizures. Herein we show that burst seizures fall almost exclusively within the short population of seizure durations; however, a short duration event was not sufficient to induce or imply bursting. We can therefore conclude that in addition to distinct mechanisms underlying seizure duration, there are separate factors regulating bursts of seizures. We show that bursts were a robust phenomenon in our patient cohort, which were consistent with overdispersed seizure rates, suggesting long-memory dynamics.

Objective Implanting subdural and penetrating electrodes in the brain causes acute trauma and inflammation that affect intracranial electroencephalographic (iEEG) recordings. This behavior and its potential impact on clinical decision-making and algorithms for implanted devices have not been assessed in detail. In this study we aim to characterize the temporal and spatial variability of continuous, prolonged human iEEG recordings. Approach Intracranial electroencephalography from 15 patients with drug-refractory epilepsy, each implanted with 16 subdural electrodes and continuously monitored for an average of 18 months, was included in this study. Time and spectral domain features were computed each day for each channel for the duration of each patient's recording. Metrics to capture post-implantation feature changes and inflexion points were computed on group and individual levels. A linear mixed model was used to characterize transient group-level changes in feature values post-implantation and independent linear models were used to describe individual variability. Main results A significant decline in features important to seizure detection and prediction algorithms (mean line length, energy, and half-wave), as well as mean power in the Berger and high gamma bands, was observed in many patients over 100 d following implantation. In addition, spatial variability across electrodes declines post-implantation following a similar timeframe. All selected features decreased by 14-50% in the initial 75 d of recording on the group level, and at least one feature demonstrated this pattern in 13 of the 15 patients. Our findings indicate that iEEG signal features demonstrate increased variability following implantation, most notably in the weeks immediately post-implant. Significance These findings suggest that conclusions drawn from iEEG, both clinically and for research, should account for spatiotemporal signal variability and that properly assessing the iEEG in patients, depending upon the application, may require extended monitoring.

Brain-computer interfaces are commonly proposed to assist individuals with locked-in syndrome to interact with the world around them. In this paper, we present a pipeline to move from recorded brain signals to real-time classification on a low-power platform, such as IBM's TrueNorth Neurosynaptic System. Our results on a EEG-based hand squeeze task show that using a convolutional neural network and a time preserving signal representation strategy provides a good balance between high accuracy and feasibility in a real-time application. This pathway can be adapted to the management of a variety of conditions, including spinal cord injury, epilepsy and Parkinson's disease.