Seizure detection using the wristband accelerometer, gyroscope, and surface electromyogram signals based on in-hospital and out-of-hospital dataset

OBJECTIVE

Wearable devices are effective for detecting generalized tonic-clonic seizures (GTCS). However, many daily activities are often misclassified as GTCS, leading to a decline in user confidence. This study recommends utilizing wristband three-axis accelerometer (ACC), three-axis gyroscope (GYRO), and surface electromyography (sEMG) signals for GTCS detection and presents a novel seizure detection algorithm that offers high sensitivity and a reduced false alarm rate (FAR).

METHODS

Inpatients with epilepsy and out-of-hospital healthy subjects were recruited and required to wear a wristband device to collect wristband signals. The proposed algorithm comprises five steps: preprocessing, motion filtering, feature extraction, classification, and postprocessing. The variations in performance across different signal combinations were compared. Additionally, the impact of training the model using only inpatient data versus the complete dataset on the algorithm's performance was also investigated.

RESULTS

Wristband signals were collected from 45 patients and 30 healthy subjects, encompassing a total of 3367.3 h and including 60 GTCS. The proposed algorithm achieved 100 % sensitivity and a FAR of 0.1070/24 h. It demonstrated higher sensitivity and lower FAR compared to combinations with fewer signal modalities. In addition, the model trained on only in-hospital data demonstrates high sensitivity (98.33 %) and high FAR (0.9845/24 h).

SIGNIFICANCE

The algorithm proposed for detecting GTCS using wristband ACC, GYRO, and sEMG signals achieved encouraging results, demonstrating the feasibility of this signal combination. Furthermore, incorporating out-of-hospital data into model training proved to be an effective solution for reducing FAR, which could facilitate the clinical application of seizure detection algorithms.

Movement Disorders and Smart Wrist Devices: A Comprehensive Study

In the medical field, there are several very different movement disorders, such as tremors, Parkinson's disease, or Huntington's disease. A wide range of motor and non-motor symptoms characterizes them. It is evident that in the modern era, the use of smart wrist devices, such as smartwatches, wristbands, and smart bracelets is spreading among all categories of people. This diffusion is justified by the limited costs, ease of use, and less invasiveness (and consequently greater acceptability) than other types of sensors used for health status monitoring. This systematic review aims to synthesize research studies using smart wrist devices for a specific class of movement disorders. Following PRISMA-S guidelines, 130 studies were selected and analyzed. For each selected study, information is provided relating to the smartwatch/wristband/bracelet model used (whether it is commercial or not), the number of end-users involved in the experimentation stage, and finally the characteristics of the benchmark dataset possibly used for testing. Moreover, some articles also reported the type of raw data extracted from the smart wrist device, the implemented designed algorithmic pipeline, and the data classification methodology. It turned out that most of the studies have been published in the last ten years, showing a growing interest in the scientific community. The selected articles mainly investigate the relationship between smart wrist devices and Parkinson's disease. Epilepsy and seizure detection are also research topics of interest, while there are few papers analyzing gait disorders, Huntington's Disease, ataxia, or Tourette Syndrome. However, the results of this review highlight the difficulties still present in the use of the smartwatch/wristband/bracelet for the identified categories of movement disorders, despite the advantages these technologies could bring in the dissemination of low-cost solutions usable directly within living environments and without the need for caregivers or medical personnel.

The Role of Wearable Devices in Chronic Disease Monitoring and Patient Care: A Comprehensive Review

Wearable health devices are becoming vital in chronic disease management because they offer real-time monitoring and personalized care. This review explores their effectiveness and challenges across medical fields, including cardiology, respiratory health, neurology, endocrinology, orthopedics, oncology, and mental health. A thorough literature search identified studies focusing on wearable devices' impact on patient outcomes. In cardiology, wearables have proven effective for monitoring hypertension, detecting arrhythmias, and aiding cardiac rehabilitation. In respiratory health, these devices enhance asthma management and continuous monitoring of critical parameters. Neurological applications include seizure detection and Parkinson's disease management, with wearables showing promising results in improving patient outcomes. In endocrinology, wearable technology advances thyroid dysfunction monitoring, fertility tracking, and diabetes management. Orthopedic applications include improved postsurgical recovery and rehabilitation, while wearables help in early complication detection in oncology. Mental health benefits include anxiety detection, post-traumatic stress disorder management, and stress reduction through wearable biofeedback. In conclusion, wearable health devices offer transformative potential for managing chronic illnesses by enhancing real-time monitoring and patient engagement. Despite significant improvements in adherence and outcomes, challenges with data accuracy and privacy persist. However, with ongoing innovation and collaboration, we can all be part of the solution to maximize the benefits of wearable technologies in healthcare.

Real-World Accuracy of Wearable Activity Trackers for Detecting Medical Conditions: Systematic Review and Meta-Analysis

BACKGROUND

Wearable activity trackers, including fitness bands and smartwatches, offer the potential for disease detection by monitoring physiological parameters. However, their accuracy as specific disease diagnostic tools remains uncertain.

OBJECTIVE

This systematic review and meta-analysis aims to evaluate whether wearable activity trackers can be used to detect disease and medical events.

METHODS

Ten electronic databases were searched for studies published from inception to April 1, 2023. Studies were eligible if they used a wearable activity tracker to diagnose or detect a medical condition or event (eg, falls) in free-living conditions in adults. Meta-analyses were performed to assess the overall area under the curve (%), accuracy (%), sensitivity (%), specificity (%), and positive predictive value (%). Subgroup analyses were performed to assess device type (Fitbit, Oura ring, and mixed). The risk of bias was assessed using the Joanna Briggs Institute Critical Appraisal Checklist for Diagnostic Test Accuracy Studies.

RESULTS

A total of 28 studies were included, involving a total of 1,226,801 participants (age range 28.6-78.3). In total, 16 (57%) studies used wearables for diagnosis of COVID-19, 5 (18%) studies for atrial fibrillation, 3 (11%) studies for arrhythmia or abnormal pulse, 3 (11%) studies for falls, and 1 (4%) study for viral symptoms. The devices used were Fitbit (n=6), Apple watch (n=6), Oura ring (n=3), a combination of devices (n=7), Empatica E4 (n=1), Dynaport MoveMonitor (n=2), Samsung Galaxy Watch (n=1), and other or not specified (n=2). For COVID-19 detection, meta-analyses showed a pooled area under the curve of 80.2% (95% CI 71.0%-89.3%), an accuracy of 87.5% (95% CI 81.6%-93.5%), a sensitivity of 79.5% (95% CI 67.7%-91.3%), and specificity of 76.8% (95% CI 69.4%-84.1%). For atrial fibrillation detection, pooled positive predictive value was 87.4% (95% CI 75.7%-99.1%), sensitivity was 94.2% (95% CI 88.7%-99.7%), and specificity was 95.3% (95% CI 91.8%-98.8%). For fall detection, pooled sensitivity was 81.9% (95% CI 75.1%-88.1%) and specificity was 62.5% (95% CI 14.4%-100%).

CONCLUSIONS

Wearable activity trackers show promise in disease detection, with notable accuracy in identifying atrial fibrillation and COVID-19. While these findings are encouraging, further research and improvements are required to enhance their diagnostic precision and applicability.

TRIAL REGISTRATION

Prospero CRD42023407867; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=407867.

Learning Robust Representations of Tonic-Clonic Seizures With Cyclic Transformer

Tonic-clonic seizures (TCSs) pose a significant risk for sudden unexpected death in epilepsy (SUDEP). Previous research has highlighted the potential of multimodal wearable seizure detection systems in accurately detecting TCSs through continuous monitoring, enabling timely alarms and potentially preventing SUDEP. However, such multimodal systems carry a higher risk of sensor malfunction. In this paper, we propose a cyclic transformer approach to address these challenges. The cyclic transformer learns a robust representation by performing circular modal translations between the source and target modalities. It leverages back-translation as regularization technique to enhance the discriminative power of the learned representation. Notably, the proposed cyclic transformer is trained on paired multimodal data but requires only a single source modality during deployment. This characteristic ensures the robustness of the cyclic transformer to perturbations or missing information in the target modality. Experimental results demonstrate that the proposed cyclic transformer achieves competitive performance compared with existing multimodal systems. While both approaches were trained using EEG and EMG data, the cyclic transformer exclusively employs EEG data for testing, diverging from the state-of-the-art's utilization of both EEG and EMG data during test. This showcases the effectiveness of the cyclic transformer in multimodal TCSs detection, offering a promising approach for enhancing the accuracy and robustness of seizure detection systems while mitigating the risks associated with sensor malfunction.

Epileptic seizure detection using CHB-MIT dataset: The overlooked perspectives

Epilepsy is a life-threatening neurological condition. Manual detection of epileptic seizures (ES) is laborious and burdensome. Machine learning techniques applied to electroencephalography (EEG) signals are widely used for automatic seizure detection. Some key factors are worth considering for the real-world applicability of such systems: (i) continuous EEG data typically has a higher class imbalance; (ii) higher variability across subjects is present in physiological signals such as EEG; and (iii) seizure event detection is more practical than random segment detection. Most prior studies failed to address these crucial factors altogether for seizure detection. In this study, we intend to investigate a generalized cross-subject seizure event detection system using the continuous EEG signals from the CHB-MIT dataset that considers all these overlooked aspects. A 5-second non-overlapping window is used to extract 92 features from 22 EEG channels; however, the most significant 32 features from each channel are used in experimentation. Seizure classification is done using a Random Forest (RF) classifier for segment detection, followed by a post-processing method used for event detection. Adopting all the above-mentioned essential aspects, the proposed event detection system achieved 72.63% and 75.34% sensitivity for subject-wise 5-fold and leave-one-out analyses, respectively. This study presents the real-world scenario for ES event detectors and furthers the understanding of such detection systems.

Machine Learning and Artificial Intelligence Applications to Epilepsy: a Review for the Practicing Epileptologist

PURPOSE OF REVIEW

Machine Learning (ML) and Artificial Intelligence (AI) are data-driven techniques to translate raw data into applicable and interpretable insights that can assist in clinical decision making. Some of these tools have extremely promising initial results, earning both great excitement and creating hype. This non-technical article reviews recent developments in ML/AI in epilepsy to assist the current practicing epileptologist in understanding both the benefits and limitations of integrating ML/AI tools into their clinical practice.

RECENT FINDINGS

ML/AI tools have been developed to assist clinicians in almost every clinical decision including (1) predicting future epilepsy in people at risk, (2) detecting and monitoring for seizures, (3) differentiating epilepsy from mimics, (4) using data to improve neuroanatomic localization and lateralization, and (5) tracking and predicting response to medical and surgical treatments. We also discuss practical, ethical, and equity considerations in the development and application of ML/AI tools including chatbots based on Large Language Models (e.g., ChatGPT). ML/AI tools will change how clinical medicine is practiced, but, with rare exceptions, the transferability to other centers, effectiveness, and safety of these approaches have not yet been established rigorously. In the future, ML/AI will not replace epileptologists, but epileptologists with ML/AI will replace epileptologists without ML/AI.

Personalized seizure signature: An interpretable approach to false alarm reduction for long-term epileptic seizure detection

OBJECTIVE

Long-term automatic detection of focal seizures remains one of the major challenges in epilepsy due to the unacceptably high number of false alarms from state-of-the-art methods. Our aim was to investigate to what extent a new patient-specific approach based on similarly occurring morphological electroencephalographic (EEG) signal patterns could be used to distinguish seizures from nonseizure events, as well as to estimate its maximum performance.

METHODS

We evaluated our approach on >5500 h of long-term EEG recordings using two public datasets: the PhysioNet.org Children's Hospital Boston-Massachusetts Institute of Technology (CHB-MIT) Scalp EEG database and the EPILEPSIAE European epilepsy database. We visually identified a set of similarly occurring morphological patterns (seizure signature) seen simultaneously over two different EEG channels, and within two randomly selected seizures from each individual. The same seizure signature was then searched for in the entire recording from the same patient using dynamic time warping (DTW) as a similarity metric, with a threshold set to reflect the maximum sensitivity our algorithm could achieve without false alarm.

RESULTS

At a DTW threshold providing no false alarm during the entire recordings, the mean seizure detection sensitivity across patients was 84%, including 96% for the CHB-MIT database and 74% for the European epilepsy database. A 100% sensitivity was reached in 50% of patients, including 79% from the CHB-MIT database and 27% from the European epilepsy database. The median latency from seizure onset to its detection was 17 ± 10 s, with 84% of seizures being detected within 40 s.

SIGNIFICANCE

Personalized EEG signature combined with DTW appears to be a promising method to detect ictal events from a limited number of EEG channels with high sensitivity despite low rate of false alarms, high degree of interpretability, and low computational complexity, compatible with its future use in wearable devices.

A Review of Commercial and Non-Commercial Wearables Devices for Monitoring Motor Impairments Caused by Neurodegenerative Diseases

Neurodegenerative diseases (NDDs) are among the 10 causes of death worldwide. The effects of NDDs, including irreversible motor impairments, have an impact not only on patients themselves but also on their families and social environments. One strategy to mitigate the pain of NDDs is to early identify and remotely monitor related motor impairments using wearable devices. Technological progress has contributed to reducing the hardware complexity of mobile devices while simultaneously improving their efficiency in terms of data collection and processing and energy consumption. However, perhaps the greatest challenges of current mobile devices are to successfully manage the security and privacy of patient medical data and maintain reasonable costs with respect to the traditional patient consultation scheme. In this work, we conclude: (1) Falls are most monitored for Parkinson's disease, while tremors predominate in epilepsy and Alzheimer's disease. These findings will provide guidance for wearable device manufacturers to strengthen areas of opportunity that need to be addressed, and (2) Of the total universe of commercial wearables devices that are available on the market, only a few have FDA approval, which means that there is a large number of devices that do not safeguard the integrity of the users who use them.

Wearable Epileptic Seizure Prediction System Based on Machine Learning Techniques Using ECG, PPG and EEG Signals

Epileptic seizures have a great impact on the quality of life of people who suffer from them and further limit their independence. For this reason, a device that would be able to monitor patients' health status and warn them for a possible epileptic seizure would improve their quality of life. With this aim, this article proposes the first seizure predictive model based on Ear EEG, ECG and PPG signals obtained by means of a device that can be used in a static and outpatient setting. This device has been tested with epileptic people in a clinical environment. By processing these data and using supervised machine learning techniques, different predictive models capable of classifying the state of the epileptic person into normal, pre-seizure and seizure have been developed. Subsequently, a reduced model based on Boosted Trees has been validated, obtaining a prediction accuracy of 91.5% and a sensitivity of 85.4%. Thus, based on the accuracy of the predictive model obtained, it can potentially serve as a support tool to determine the status epilepticus and prevent a seizure, thereby improving the quality of life of these people.

Seizures detection using multimodal signals: a scoping review

Introduction. Epileptic seizures are common neurological disorders in the world, impacting 65 million people globally. Around 30% of patients with seizures suffer from refractory epilepsy, where seizures are not controlled by medications. The unpredictability of seizures makes it essential to have a continuous seizure monitoring system outside clinical settings for the purpose of minimizing patients' injuries and providing additional pathways for evaluation and treatment follow-up. Autonomic changes related to seizure events have been extensively studied and attempts made to apply them for seizure detection and prediction tasks. This scoping review aims to depict current research activities associated with the implementation of portable, wearable devices for seizure detection or prediction and inform future direction in continuous seizure tracking in ambulatory settings.Methods. Overall methodology framework includes 5 essential stages: research questions identification, relevant studies identification, selection of studies, data charting and summarizing the findings. A systematic searching strategy guided by systematic reviews and meta-analysis (PRISMA) was implemented to identify relevant records on two databases (PubMed, IEEE).Results. A total of 30 articles were included in our final analysis. Most of the studies were conducted off-line and employed consumer-graded wearable device. ACM is the dominant modality to be used in seizure detection, and widely deployed algorithms entail Support Vector Machine, Random Forest and threshold-based approach. The sensitivity ranged from 33.2% to 100% for single modality with a false alarm rate (FAR) ranging from 0.096 to 14.8 d-1. Multimodality has a sensitivity ranging from 51% to 100% with FAR ranging from 0.12 to 17.7 d-1.Conclusion. The overall performance in seizure detection system based on non-cerebral physiological signals is promising, especially for the detection of motor seizures and seizures accompanied with intense ictal autonomic changes.

Detection of Epileptic Seizure Using Accelerometer Time Series Data and Hidden Markov Model

Epilepsy is one of the most prevalent neurological diseases globally, which causes seizures in the patient. As per a survey done worldwide, it is found that approximately 70 million people are living with epilepsy (~1% of the total population of the world). Effective detection of these seizures requires specialized approaches such as video and electroencephalography monitoring, which are expensive and are mainly available at specialized hospitals and institutes. Hence, there is a need to develop simpler and affordable systems that can be made available to health care centers and patients for accurate detection of epileptic seizures. A wireless remote monitoring system based on a wrist-worn accelerometer is an optimum choice for the same. Sophisticated algorithms need to be developed for effectively detecting seizure events from this accelerometer data with minimal false alarms. This paper presents a Hidden Markov Model (HMM) based probabilistic approach applied to the reduced-dimension feature vector representation of time-series accelerometer data to detect epileptic seizures. The results obtained from the HMM were compared with three commonly used machine learning models viz. support vector machine (SVM), logistic regression, and random forest. The proposed approach was able to detect 95.7% of seizures with a low false alarm rate of 14.8% with a run time of just under 24 seconds.

Assessing the Feasibility of Acoustic Based Seizure Detection

OBJECTIVE

Long-term monitoring of epilepsy patients outside of hospital settings is impractical due to the complexity and costs associated with electroencephalogram (EEG) systems. Alternative sensing modalities that can acquire, and automatically interpret signals through easy-to-use wearable devices, are needed to help with at-home management of the disease. In this paper, a novel machine learning algorithm is presented for detecting epileptic seizures using acoustic physiological signals acquired from the neck using a wearable device.

METHODS

Acoustic signals from an existing database, were processed, to extract their Mel-frequency Cepstral Coefficients (MFCCs) which were used to train RUSBoost classifiers to identify ictal and non-ictal acoustic segments. A postprocessing stage was then applied to the segment classification results to identify seizures episodes.

RESULTS

Tested on 667 hours of acoustic data acquired from 15 patients with at least one seizure, the algorithm achieved a detection sensitivity of 88.1% (95% CI: 79%-97%) from a total of 36 seizures, out of which 24 had no motor manifestations, with a FPR of 0.83/h, and a median detection latency of -42 s.

CONCLUSION

The results demonstrated for the first time the ability to identify seizures using acoustic internal body signals acquired on the neck.

SIGNIFICANCE

The results of this paper validate the feasibility of using internal physiological sounds for seizure detection, which could potentially be of use for the development of novel, wearable, very simple to use, long term monitoring, or seizure detection systems; circumventing the practical limitations of EEG monitoring outside hospital settings, or systems based on sensing modalities that work on convulsive seizures only.

Automated seizure detection with non-invasive wearable devices: A systematic review and meta-analysis

Objective

This study was undertaken to review the reported performance of noninvasive wearable devices in detecting epileptic seizures and psychogenic nonepileptic seizures (PNES).

Methods

We conducted a systematic review and meta‐analysis of studies reported up to November 15, 2021. We included studies that used video‐electroencephalographic (EEG) monitoring as the gold standard to determine the sensitivity and false alarm rate (FAR) of noninvasive wearables for automated seizure detection.

Results

Twenty‐eight studies met the criteria for the systematic review, of which 23 were eligible for meta‐analysis. These studies (1269 patients in total, median recording time = 52.9 h per patient) investigated devices for tonic–clonic seizures using wrist‐worn and/or ankle‐worn devices to measure three‐dimensional accelerometry (15 studies), and/or wearable surface devices to measure electromyography (eight studies). The mean sensitivity for detecting tonic–clonic seizures was .91 (95% confidence interval [CI] = .85–.96, I² = 83.8%); sensitivity was similar between the wrist‐worn (.93) and surface devices (.90). The overall FAR was 2.1/24 h (95% CI = 1.7–2.6, I² = 99.7%); FAR was higher in wrist‐worn (2.5/24 h) than in wearable surface devices (.96/24 h). Three of the 23 studies also detected PNES; the mean sensitivity and FAR from these studies were 62.9% and .79/24 h, respectively. Four studies detected both focal and tonic–clonic seizures, and one study detected focal seizures only; the sensitivities ranged from 31.1% to 93.1% in these studies.

Significance

Reported noninvasive wearable devices had high sensitivity but relatively high FARs in detecting tonic–clonic seizures during limited recording time in a video‐EEG setting. Future studies should focus on reducing FAR, detection of other seizure types and PNES, and longer recording in the community.

Intra- and Inter-Subject Perspectives on the Detection of Focal Onset Motor Seizures in Epilepsy Patients

Focal onset epileptic seizures are highly heterogeneous in their clinical manifestations, and a robust seizure detection across patient cohorts has to date not been achieved. Here, we assess and discuss the potential of supervised machine learning models for the detection of focal onset motor seizures by means of a wrist-worn wearable device, both in a personalized context as well as across patients. Wearable data were recorded in-hospital from patients with epilepsy at two epilepsy centers. Accelerometry, electrodermal activity, and blood volume pulse data were processed and features for each of the biosignal modalities were calculated. Following a leave-one-out approach, a gradient tree boosting machine learning model was optimized and tested in an intra-subject and inter-subject evaluation. In total, 20 seizures from 9 patients were included and we report sensitivities of 67% to 100% and false alarm rates of down to 0.85 per 24 h in the individualized assessment. Conversely, for an inter-subject seizure detection methodology tested on an out-of-sample data set, an optimized model could only achieve a sensitivity of 75% at a false alarm rate of 13.4 per 24 h. We demonstrate that robustly detecting focal onset motor seizures with tonic or clonic movements from wearable data may be possible for individuals, depending on specific seizure manifestations.

Automatic annotation correction for wearable EEG based epileptic seizure detection

OBJECTIVE

Video-electroencephalography (vEEG), which defines the ground truth for the detection of epileptic seizures, is inadequate for long-term home monitoring. Thanks to their advantages in comfort and unobtrusiveness, wearable EEG devices have been suggested as a solution for home monitoring. However, one of the challenges in data-driven automated seizure detection with wearable EEG data is to have reliable seizure annotations. Seizure annotations on the gold-standard 25-channel vEEG recordings may not be optimal to delineate seizure activity on the concomitantly recorded wearable EEG, due to artifacts or absence of ictal activity on the limited set of electrodes of the wearable EEG. This paper aims to develop an automatic approach to correct the imperfect annotations of seizure activity on wearable EEG, which can be used to train seizure detection algorithms.

APPROACH

This paper first investigates the effectiveness of correcting the seizure annotations for the training set with a visual annotation correction. Then a novel approach has been proposed to automatically remove non-seizure data from wearable EEG in epochs annotated as seizures in gold-standard video-EEG recordings. The performance of the automatic annotation correction approach was evaluated by comparing the seizure detection models trained with 1. original vEEG seizure annotations, 2. visually corrected seizure annotations, and 3. automatically corrected seizure annotations.

RESULTS

The automatic seizure detection approach trained with automatically corrected seizure annotations was more sensitive and had fewer false-positive detections compared to the approach trained with visually corrected seizure annotations, and the approach trained with the original seizure annotations from gold-standard vEEG.

SIGNIFICANCE

The wearable EEG seizure detection approach performs better when trained with automatic seizure annotation correction.

Identification of Ictal Tachycardia in Focal Motor- and Non-Motor Seizures by Means of a Wearable PPG Sensor

Photoplethysmography (PPG) as an additional biosignal for a seizure detector has been underutilized so far, which is possibly due to its susceptibility to motion artifacts. We investigated 62 focal seizures from 28 patients with electrocardiography-based evidence of ictal tachycardia (IT). Seizures were divided into subgroups: those without epileptic movements and those with epileptic movements not affecting and affecting the extremities. PPG-based heart rate (HR) derived from a wrist-worn device was calculated for sections with high signal quality, which were identified using spectral entropy. Overall, IT based on PPG was identified in 37 of 62 (60%) seizures (9/19, 7/8, and 21/35 in the three groups, respectively) and could be found prior to the onset of epileptic movements affecting the extremities in 14/21 seizures. In 30/37 seizures, PPG-based IT was in good temporal agreement (<10 s) with ECG-based IT, with an average delay of 5.0 s relative to EEG onset. In summary, we observed that the identification of IT by means of a wearable PPG sensor is possible not only for non-motor seizures but also in motor seizures, which is due to the early manifestation of IT in a relevant subset of focal seizures. However, both spontaneous and epileptic movements can impair PPG-based seizure detection.

Video-Based Detection of Generalized Tonic-Clonic Seizures Using Deep Learning

Timely detection of seizures is crucial to implement optimal interventions, and may help reduce the risk of sudden unexpected death in epilepsy (SUDEP) in patients with generalized tonic-clonic seizures (GTCSs). While video-based automated seizure detection systems may be able to provide seizure alarms in both in-hospital and at-home settings, earlier studies have primarily employed hand-designed features for such a task. In contrast, deep learning-based approaches do not rely on prior feature selection and have demonstrated outstanding performance in many data classification tasks. Despite these advantages, neural network-based video classification has rarely been attempted for seizure detection. We here assessed the feasibility and efficacy of automated GTCSs detection from videos using deep learning. We retrospectively identified 76 GTCS videos from 37 participants who underwent long-term video-EEG monitoring (LTM) along with interictal video data from the same patients, and 10 full-night seizure-free recordings from additional patients. Using a leave-one-subject-out cross-validation approach (LOSO-CV), we evaluated the performance to detect seizures based on individual video frames (convolutional neural networks, CNNs) or video sequences [CNN+long short-term memory (LSTM) networks]. CNN+LSTM networks based on video sequences outperformed GTCS detection based on individual frames yielding a mean sensitivity of 88% and mean specificity of 92% across patients. The average detection latency after presumed clinical seizure onset was 22 seconds. Detection performance increased as a function of training dataset size. Collectively, we demonstrated that automated video-based GTCS detection with deep learning is feasible and efficacious. Deep learning-based methods may be able to overcome some limitations associated with traditional approaches using hand-crafted features, serve as a benchmark for future methods and analyses, and improve further with larger datasets.

A machine learning model for multi-night actigraphic detection of chronic insomnia: development and validation of a pre-screening tool

We propose a novel machine learning-based method for analysing multi-night actigraphy signals to objectively classify and differentiate nocturnal awakenings in individuals with chronic insomnia (CI) and their cohabiting healthy partners. We analysed nocturnal actigraphy signals from 40 cohabiting couples with one partner seeking treatment for insomnia. We extracted 12 time-domain dynamic and nonlinear features from the actigraphy signals to classify nocturnal awakenings in healthy individuals and those with CI. These features were then used to train two machine learning classifiers, random forest (RF) and support vector machine (SVM). An optimization algorithm that incorporated the predicted quality of each night for each individual was used to classify individuals into CI or healthy sleepers. Using the proposed actigraphic signal analysis technique, coupled with a rigorous leave-one-out validation approach, we achieved a classification accuracy of 80% (sensitivity: 76%, specificity: 82%) in classifying CI individuals and their healthy bed partners. The RF classifier (accuracy: 80%) showed a better performance than SVM (accuracy: 75%). Our approach to analysing the multi-night nocturnal actigraphy recordings provides a new method for screening individuals with CI, using wrist-actigraphy devices, facilitating home monitoring.

Non-invasive wearable seizure detection using long-short-term memory networks with transfer learning

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.

Automated seizure detection using wearable devices: A clinical practice guideline of the International League Against Epilepsy and the International Federation of Clinical Neurophysiology

The objective of this clinical practice guideline (CPG) is to provide recommendations for healthcare personnel working with patients with epilepsy on the use of wearable devices for automated seizure detection in patients with epilepsy, in outpatient, ambulatory settings. The Working Group of the International League Against Epilepsy (ILAE) and the International Federation of Clinical Neurophysiology (IFCN) developed the CPG according to the methodology proposed by the ILAE Epilepsy Guidelines Working Group. We reviewed the published evidence using The Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) statement and evaluated the evidence and formulated the recommendations following the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system. We found high level of evidence for the accuracy of automated detection of generalized tonic-clonic seizures (GTCS) and focal-to-bilateral tonic-clonic seizures (FBTCS) and recommend the use of wearable automated seizure detection devices for selected patients when accurate detection of GTCS and FBTCS is recommended as a clinical adjunct. We also found a moderate level of evidence for seizure types without GTCS or FBTCS. However, it was uncertain whether the detected alarms resulted in meaningful clinical outcomes for the patients. We recommend using clinically validated devices for automated detection of GTCS and FBTCS, especially in unsupervised patients, where alarms can result in rapid intervention (weak/conditional recommendation). At present, we do not recommend clinical use of the currently available devices for other seizure types (weak/conditional recommendation). Further research and development are needed to improve the performance of automated seizure detection and to document their accuracy and clinical utility.

Automated seizure detection using wearable devices: A clinical practice guideline of the International League Against Epilepsy and the International Federation of Clinical Neurophysiology

The objective of this clinical practice guideline (CPG) is to provide recommendations for healthcare personnel working with patients with epilepsy, on the use of wearable devices for automated seizure detection in patients with epilepsy, in outpatient, ambulatory settings. The Working Group of the International League Against Epilepsy and the International Federation of Clinical Neurophysiology developed the CPG according to the methodology proposed by the ILAE Epilepsy Guidelines Working Group. We reviewed the published evidence using The Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) statement and evaluated the evidence and formulated the recommendations following the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system. We found high level of evidence for the accuracy of automated detection of generalized tonic-clonic seizures (GTCS) and focal-to-bilateral tonic-clonic seizures (FBTCS) and recommend use of wearable automated seizure detection devices for selected patients when accurate detection of GTCS and FBTCS is recommended as a clinical adjunct. We also found moderate level of evidence for seizure types without GTCs or FBTCs. However, it was uncertain whether the detected alarms resulted in meaningful clinical outcomes for the patients. We recommend using clinically validated devices for automated detection of GTCS and FBTCS, especially in unsupervised patients, where alarms can result in rapid intervention (weak/conditional recommendation). At present, we do not recommend clinical use of the currently available devices for other seizure types (weak/conditional recommendation). Further research and development are needed to improve the performance of automated seizure detection and to document their accuracy and clinical utility.

Factors associated with delay to video-EEG in dissociative seizures

PURPOSE

While certain clinical factors suggest a diagnosis of dissociative seizures (DS), otherwise known as functional or psychogenic nonepileptic seizures (PNES), ictal video-electroencephalography monitoring (VEM) is the gold standard for diagnosis. Diagnostic delays were associated with worse quality of life and more seizures, even after treatment. To understand why diagnoses were delayed, we evaluated which factors were associated with delay to VEM.

METHODS

Using data from 341 consecutive patients with VEM-documented dissociative seizures, we used multivariate log-normal regression with recursive feature elimination (RFE) and multiple imputation of some missing data to evaluate which of 76 clinical factors were associated with time from first dissociative seizure to VEM.

RESULTS

The mean delay to VEM was 8.4 years (median 3 years, IQR 1-10 years). In the RFE multivariate model, the factors associated with longer delay to VEM included more past antiseizure medications (0.19 log-years/medication, standard error (SE) 0.05), more medications for other medical conditions (0.06 log-years/medication, SE 0.03), history of physical abuse (0.75 log-years, SE 0.27), and more seizure types (0.36 log-years/type, SE 0.11). Factors associated with shorter delay included active employment or student status (-1.05 log-years, SE 0.21) and higher seizure frequency (0.14 log-years/log[seizure/month], SE 0.06).

CONCLUSIONS

Patients with greater medical and seizure complexity had longer delays. Delays in multiple domains of healthcare can be common for victims of physical abuse. Unemployed and non-student patients may have had more barriers to access VEM. These results support earlier referral of complex cases to a comprehensive epilepsy center.

Reliability of additional reported seizure manifestations to identify dissociative seizures

PURPOSE

Descriptions of seizure manifestations (SM), or semiology, can help localize the symptomatogenic zone and subsequently included brain regions involved in epileptic seizures, as well as identify patients with dissociative seizures (DS). Patients and witnesses are not trained observers, so these descriptions may vary from expert review of seizure video recordings of seizures. To better understand how reported factors can help identify patients with DS or epileptic seizures (ES), we evaluated the associations between more than 30 SMs and diagnosis using standardized interviews.

METHODS

Based on patient- and observer-reported data from 490 patients with diagnoses documented by video-electoencephalography, we compared the rate of each SM in five mutually exclusive groups: epileptic seizures (ES), DS, physiologic seizure-like events (PSLE), mixed DS and ES, and inconclusive testing.

RESULTS

In addition to SMs that we described in a prior manuscript, the following were associated with DS: light triggers, emotional stress trigger, pre-ictal and post-ictal headache, post-ictal muscle soreness, and ictal sensory symptoms. The following were associated with ES: triggered by missing medication, aura of déjà vu, and leftward eye deviation. There were numerous manifestations separately associated with mixed ES and DS.

CONCLUSIONS

Reported SM can help identify patients with DS, but no manifestation is pathognomonic for either ES or DS. Patients with mixed ES and DS reported factors divergent from both ES-alone and DS-alone.

Wearable seizure detection devices in refractory epilepsy

Epilepsy affects 50 million patients and their caregivers worldwide. Devices that facilitate the detection of seizures can have a large influence on a patient's quality of life, therapeutic decisions and the conduct of clinical trials with anti-epileptic drugs. This article provides an up-to-date overview and comparison between wearable seizure detection devices (WSDDs), taking into account the newly proposed standards for testing and clinical validation of devices. 16 devices were included in our comparison. The F1-score, combining the device's accurate recall and precision, was calculated for each of these devices and used to evaluate their performance. The devices were separated by development phase and ranked by F1-score from highest to lowest. We describe 16 WSDDs: 6 of which were accelerometry (ACM)-based, 3 surface electromyography-based, 1 was a wearable application of EEG, 4 had multimodal sensors and 2 other types of sensors. We observed a significant inconsistency in the description of performance measures. The devices in the most advanced development phase with the highest F1-scores incorporated ACM- and sEMG-based sensors to detect tonic-clonic seizures. This review highlights the importance of implementing standards for an optimal comparison and, therefore, improving the research and development of WSDDs. WSDDs can improve the patient's care and quality of life, decrease seizure underreporting and they could potentially prevent sudden-unexpected-death in epilepsy. We discuss the central role of the neurologist in the use of WSDDs, and why a business to business to consumer model is better than the current business to consumer model of most WSDDs.

The utility of an automated and ambulatory device for detecting and differentiating epileptic and psychogenic non-epileptic seizures

OBJECTIVE

Accurate differentiation between epileptic seizures (ES) and psychogenic non-epileptic seizures (PNES) can be challenging based on history alone. Inpatient video EEG monitoring (VEM) is often needed for a definitive diagnosis. However, VEM is highly resource intensive, is of limited availability, and cannot be undertaken over long periods. Previous research has shown that time-frequency analysis of accelerometer data could be utilized to differentiate between ES and PNES. Using a seizure detection and classification algorithm, we sought to examine the diagnostic utility of an automated analysis with an ambulatory accelerometer.

METHODS

A wrist-worn device was used to collect accelerometer data from patients during VEM admission, for diagnostic evaluation of convulsive seizures. An automated process, that involved the use of K-means clustering and support vector machines, was used to detect and classify each seizure as ES or PNES. The results were compared with VEM diagnoses determined by epileptologists blinded to the accelerometer data.

RESULTS

Twenty-four convulsive seizures, consisting of at least 20 seconds of sustained continuous activity, recorded from 11 patients during inpatient VEM (13 PNES from five patients and 11 ES from six patients) were included for analysis. The automated system detected all convulsive seizures (ES, PNES) from >661 hours of recording with 67 false alarms (2.4 per 24 hours). The sensitivity and specificity for classifying ES from PNES were 72.7% and 100%, respectively. The positive and negative predictive values for classifying PNES were 81.3% and 100%, respectively. There was no significant difference between the classification results obtained from the automation process and the VEM diagnoses.

SIGNIFICANCE

This automated system can potentially provide a wearable out-of-hospital seizure diagnostic monitoring system.

Novel features for capturing temporal variations of rhythmic limb movement to distinguish convulsive epileptic and psychogenic nonepileptic seizures

OBJECTIVE

To investigate the characteristics of motor manifestation during convulsive epileptic and psychogenic nonepileptic seizures (PNES), captured using a wrist-worn accelerometer (ACM) device. The main goal was to find quantitative ACM features that can differentiate between convulsive epileptic and convulsive PNES.

METHODS

In this study, motor data were recorded using wrist-worn ACM-based devices. A total of 83 clinical events were recorded: 39 generalized tonic-clonic seizures (GTCS) from 12 patients with epilepsy, and 44 convulsive PNES from 7 patients (one patient had both GTCS and PNES). The temporal variations in the ACM traces corresponding to 39 GTCS and 44 convulsive PNES events were extracted using Poincaré maps. Two new indices-tonic index (TI) and dispersion decay index (DDI)-were used to quantify the Poincaré-derived temporal variations for every GTCS and convulsive PNES event.

RESULTS

The TI and DDI of Poincaré-derived temporal variations for GTCS events were higher in comparison to convulsive PNES events (P < 0.001). The onset and the subsiding patterns captured by TI and DDI differentiated between epileptic and convulsive nonepileptic seizures. An automated classifier built using TI and DDI of Poincaré-derived temporal variations could correctly differentiate 42 (sensitivity: 95.45%) of 44 convulsive PNES events and 37 (specificity: 94.87%) of 39 GTCS events. A blinded review of the Poincaré-derived temporal variations in GTCS and convulsive PNES by epileptologists differentiated 26 (sensitivity: 70.27%) of 44 PNES events and 33 (specificity: 86.84%) of 39 GTCS events correctly.

SIGNIFICANCE

In addition to quantifying the motor manifestation mechanism of GTCS and convulsive PNES, the proposed approach also has diagnostic significance. The new ACM features incorporate clinical characteristics of GTCS and PNES, thus providing an accurate, low-cost, and practical alternative to differential diagnosis of PNES.