A rule-based seizure prediction method for focal neocortical epilepsy

Spectral Slope and Lempel-Ziv Complexity as Robust Markers of Brain States during Sleep and Wakefulness

Nonoscillatory measures of brain activity such as the spectral slope and Lempel-Ziv complexity are affected by many neurological disorders and modulated by sleep. A multitude of frequency ranges, particularly a broadband (encompassing the full spectrum) and a narrowband approach, have been used especially for estimating the spectral slope. However, the effects of choosing different frequency ranges have not yet been explored in detail. Here, we evaluated the impact of sleep stage and task engagement (resting, attention, and memory) on slope and complexity in a narrowband (30-45 Hz) and broadband (1-45 Hz) frequency range in 28 healthy male human subjects (21.54 ± 1.90 years) using a within-subject design over 2 weeks with three recording nights and days per subject. We strived to determine how different brain states and frequency ranges affect slope and complexity and how the two measures perform in comparison. In the broadband range, the slope steepened, and complexity decreased continuously from wakefulness to N3 sleep. REM sleep, however, was best discriminated by the narrowband slope. Importantly, slope and complexity also differed between tasks during wakefulness. While narrowband complexity decreased with task engagement, the slope flattened in both frequency ranges. Interestingly, only the narrowband slope was positively correlated with task performance. Our results show that slope and complexity are sensitive indices of brain state variations during wakefulness and sleep. However, the spectral slope yields more information and could be used for a greater variety of research questions than Lempel-Ziv complexity, especially when a narrowband frequency range is used.

Research progress on the electrophysiological indicators to predict the efficacy of vagus nerve stimulation for drug-refractory epilepsy

Vagus nerve stimulation (VNS) is an important treatment option for drug-refractory epilepsy (DRE), with well-established efficacy and safety in clinical practice for more than 20 years. However, it is very difficult to find the optimal electrophysiological indicators for the effectiveness of VNS on DRE because the mechanism of action is unknown. In this review, we provide an update of the potential applications of VNS outcomes in patients with drug-resistant epilepsy. Electroencephalographic (EEG) activity, event-related potentials, EEG synchronization levels, magnetoencephalographic, laryngeal muscle evoked potentials, and heart rate variability are potential biomarkers for VNS outcomes in people with DRE.

An approach to detect and predict epileptic seizures with high accuracy using convolutional neural networks and single-lead-ECG signal

One of the epileptic patients' challenges is to detect the time of seizures and the possibility of predicting. This research aims to provide an algorithm based on deep learning to detect and predict the time of seizure from one to two minutes before its occurrence. The proposed Convolutional Neural Network (CNN) can detect and predict the occurrence of focal epilepsy seizures through single-lead-ECG signal processing instead of using EEG signals. The structure of the proposed CNN for seizure detection and prediction is the same. Considering the requirements of a wearable system, after a few light pre-processing steps, the ECG signal can be used as input to the neural network without any manual feature extraction step. The desired neural network learns purposeful features according to the labelled ECG signals and then performs the classification of these signals. Training of 39-layer CNN for seizure detection and prediction has been done separately. The proposed method can detect seizures with an accuracy of 98.84% and predict them with an accuracy of 94.29%. With this approach, the ECG signal can be a promising indicator for the construction of portable systems for monitoring the status of epileptic patients.

An overview of machine learning and deep learning techniques for predicting epileptic seizures

Epilepsy is a neurological disorder (the third most common, following stroke and migraines). A key aspect of its diagnosis is the presence of seizures that occur without a known cause and the potential for new seizures to occur. Machine learning has shown potential as a cost-effective alternative for rapid diagnosis. In this study, we review the current state of machine learning in the detection and prediction of epileptic seizures. The objective of this study is to portray the existing machine learning methods for seizure prediction. Internet bibliographical searches were conducted to identify relevant literature on the topic. Through cross-referencing from key articles, additional references were obtained to provide a comprehensive overview of the techniques. As the aim of this paper aims is not a pure bibliographical review of the subject, the publications here cited have been selected among many others based on their number of citations. To implement accurate diagnostic and treatment tools, it is necessary to achieve a balance between prediction time, sensitivity, and specificity. This balance can be achieved using deep learning algorithms. The best performance and results are often achieved by combining multiple techniques and features, but this approach can also increase computational requirements.

Seizure detection with reduced electroencephalogram channels: research trends and outlook

Epilepsy is a prevalent condition characterized by recurrent, unpredictable seizures. Monitoring with surface electroencephalography (EEG) is the gold standard for diagnosing epilepsy, but a time-consuming, uncomfortable and sometimes ineffective process for patients. Further, using EEG over a brief monitoring period has variable success, dependent on patient tolerance and seizure frequency. The availability of hospital resources and hardware and software specifications inherently restrict the options for comfortable, long-term data collection, resulting in limited data for training machine-learning models. This mini-review examines the current patient journey, providing an overview of the current state of EEG monitoring with reduced electrodes and automated channel reduction methods. Opportunities for improving data reliability through multi-modal data fusion are suggested. We assert the need for further research in electrode reduction to advance brain monitoring solutions towards portable, reliable devices that simultaneously offer patient comfort, perform ultra-long-term monitoring and expedite the diagnosis process.

Seizure Detection and Prediction by Parallel Memristive Convolutional Neural Networks

During the past two decades, epileptic seizure detection and prediction algorithms have evolved rapidly. However, despite significant performance improvements, their hardware implementation using conventional technologies, such as Complementary Metal-Oxide-Semiconductor (CMOS), in power and area-constrained settings remains a challenging task; especially when many recording channels are used. In this paper, we propose a novel low-latency parallel Convolutional Neural Network (CNN) architecture that has between 2-2,800x fewer network parameters compared to State-Of-The-Art (SOTA) CNN architectures and achieves 5-fold cross validation accuracy of 99.84% for epileptic seizure detection, and 99.01% and 97.54% for epileptic seizure prediction, when evaluated using the University of Bonn Electroencephalogram (EEG), CHB-MIT and SWEC-ETHZ seizure datasets, respectively. We subsequently implement our network onto analog crossbar arrays comprising Resistive Random-Access Memory (RRAM) devices, and provide a comprehensive benchmark by simulating, laying out, and determining hardware requirements of the CNN component of our system. We parallelize the execution of convolution layer kernels on separate analog crossbars to enable 2 orders of magnitude reduction in latency compared to SOTA hybrid Memristive-CMOS Deep Learning (DL) accelerators. Furthermore, we investigate the effects of non-idealities on our system and investigate Quantization Aware Training (QAT) to mitigate the performance degradation due to low Analog-to-Digital Converter (ADC)/Digital-to-Analog Converter (DAC) resolution. Finally, we propose a stuck weight offsetting methodology to mitigate performance degradation due to stuck [Formula: see text] memristor weights, recovering up to 32% accuracy, without requiring retraining. The CNN component of our platform is estimated to consume approximately 2.791 W of power while occupying an area of 31.255 mm² in a 22 nm FDSOI CMOS process.

Tiny CNN for Seizure Prediction in Wearable Biomedical Devices

Epilepsy is a life-threatening disease affecting millions of people all over the world. Artificial intelligence epileptic predictors offer excellent potential to improve epilepsy therapy. Particularly, deep learning models such as convolutional neural networks (CNN) can be used to accurately detect ictogenesis through deep structured learning representations. In this work, a tiny one-dimensional stacked convolutional neural network (1DSCNN) is proposed based on short-time Fourier transform (STFT) to predict epileptic seizure. The results demonstrate that the proposed method obtains better performance compared to recent state-of-the-art methods, achieving an average sensitivity of 94.44%, average false prediction rate (FPR) of 0.011/h and average area under the curve (AUC) of 0.979 on the test set of the American Epilepsy Society Seizure Prediction Challenge dataset, while featuring a model size of only 21.32kB. Furthermore, after adapting the model to 4-bit quantization, its size is significantly decreased by 7.08x with only 0.51% AUC score precision loss, which shows excellent potential for hardware-friendly wearable implementation.

Epileptic electroencephalography classification using Embedded Dynamic Mode Decomposition

OBJECTIVE

Seizure prediction devices for drug-resistant epileptic patients could lead to improved quality of life and new treatment options, but current approaches to classification of electroencephalography (EEG) segments for early identification of the pre-seizure state typically require many features and complex classifiers. We therefore propose a novel spatio-temporal EEG feature set that significantly aids in separation and easy classification of the interictal and preictal states.

APPROACH

We derive key spectral features from the Embedded Dynamic Mode Decomposition (EmDMD) of the brain state system. This method linearizes the complex spatio-temporal dynamics of the system, describing the dynamics in terms of a spectral basis of modes and eigenvalues. The relative subband spectral power and mean phase locking values of these modes prove to be good indicators of the preictal state that precedes seizure onset.

MAIN RESULTS

We analyze the linear separability and classification of preictal and interictal states based on our proposed features using seizure data extracted from the CHB-MIT scalp EEG and Kaggle American Epilepsy Society Seizure Prediction Challenge intracranial EEG databases. With a light-weight support vector machine or random forest classifier trained on these features, we classify the preictal state with a sensitivity of up to 92% and specificity of up to 89%.

SIGNIFICANCE

The EmDMD-derived features separate the preictal and interictal states, improving classification accuracy and motivating further work to incorporate them into seizure prediction algorithms.

A novel consistency-based training strategy for seizure prediction

BACKGROUND

Early prediction of epilepsy seizures can warn the patients to take precautions and improve their lives significantly. In recent years, deep learning has become increasingly predominant in seizure prediction for its outstanding performance. With the aim of predicting unseen seizures, it is essential to guarantee the generalization ability of the model, especially considering the non-stationary nature of EEG and the scarcity of seizure events in EEG recordings. Stability training against extra perturbations is an intuitive and effective way to improve the model's ability to generalize. Though a great number of deep learning methods have been developed for seizure prediction, their strategies to increase generalization performance focus on improving the model's architecture itself, and few of them pay attention to the stability of the model against small perturbations.

NEW METHOD

In this study, we propose a novel consistency-based training strategy to address this issue. The proposed strategy underlines that a robust model should maintain consistent results for the same input under extra perturbations. Specifically, during training, we use stochastic augmentations to make the input vary from iteration to iteration and consider the output as a stochastic variable. Then a consistency constraint is constructed to penalize the difference between the current output and previous outputs. In this way, the generalization ability of the model will be fully enhanced.

RESULTS

To better verify the effectiveness of our proposed strategy, we implement it in two state-of-the-art models with public-available codes, including STFT CNN and Multi-view CNN. Notably, we compare with the first baseline on a scalp EEG dataset and the other on an intracranial EEG dataset. The results show that our strategy could improve the performance significantly for both of them.

COMPARISON WITH EXISTING METHODS

Our strategy has increased the sensitivity by 7.1% and reduced the false prediction rate by 0.12/h on the first baseline while improving the AUC by 0.020 on the second baseline.

CONCLUSIONS

This study is easy to implement, providing a new solution to enhance the performance of seizure prediction.

Neuromorphic Architecture Accelerated Automated Seizure Detection in Multi-Channel Scalp EEG

Epileptic focal seizures can be localized in the brain using tracer injections during or immediately after the incidence of a seizure. A real-time automated seizure detection system with minimal latency can help time the injection properly to find the seizure origin accurately. Reliable real-time seizure detection systems have not been clinically reported yet. We developed an anomaly detection-based automated seizure detection system, using scalp-electroencephalogram (EEG) data, which can be trained using a few seizure sessions, and implemented it on commercially available hardware with parallel, neuromorphic architecture—the NeuroStack. We extracted nonlinear, statistical, and discrete wavelet decomposition features, and we developed a graphical user interface and traditional feature selection methods to select the most discriminative features. We investigated Reduced Coulomb Energy (RCE) networks and K-Nearest Neighbors (k-NN) for its several advantages, such as fast learning no local minima problem. We obtained a maximum sensitivity of 91.14%±1.77% and a specificity of 98.77%±0.57% with 5 s epoch duration. The system’s latency was 12 s, which is within most seizure event windows, which last for an average duration of 60 s. Our results showed that the CD feature consumes large computation resources and excluding it can reduce the latency to 3.6 s but at the cost of lower performance 80% sensitivity and 97% specificity. We demonstrated that the proposed methodology achieves a high specificity and an acceptable sensitivity within a short delay. Our results indicated also that individual-based RCE are superior to population-based RCE. The proposed RCE networks has been compared to SVM and ANN as a baseline for comparison as they are the most common machine learning seizure detection methods. SVM and ANN-based systems were trained on the same data as RCE and K-NN with features optimized specifically for them. RCE nets are superior to SVM and ANN. The proposed model also achieves comparable performance to the state-of-the-art deep learning techniques while not requiring a sizeable database, which is often expensive to build. These numbers indicate that the system is viable as a trigger mechanism for tracer injection.

Semisupervised Seizure Prediction in Scalp EEG Using Consistency Regularization

Early prediction of epilepsy seizures can warn the patients to take precautions and improve their lives significantly. In recent years, deep learning has become increasingly predominant in seizure prediction. However, existing deep learning-based approaches in this field require a great deal of labeled data to guarantee performance. At the same time, labeling EEG signals does require the expertise of an experienced pathologist and is incredibly time-consuming. To address this issue, we propose a novel Consistency-based Semisupervised Seizure Prediction Model (CSSPM), where only a fraction of training data is labeled. Our method is based on the principle of consistency regularization, which underlines that a robust model should maintain consistent results for the same input under extra perturbations. Specifically, by using stochastic augmentation and dropout, we consider the entire neural network as a stochastic model and apply a consistency constraint to penalize the difference between the current prediction and previous predictions. In this way, unlabeled data could be fully utilized to improve the decision boundary and enhance prediction performance. Compared with existing studies requiring all training data to be labeled, the proposed method only needs a small portion of data to be labeled while still achieving satisfactory results. Our method provides a promising solution to alleviate the labeling cost for real-world applications.

Visual Object Recognition from Single-Trial EEG Signals using Machine Learning Wrapper Techniques

Responses of the human brain to different visual stimuli elicit specific patterns in electroencephalography (EEG) signals. It is confirmed that by analyzing these patterns, we can recognize the category of the visited objects. However, high levels of noise and artifacts in EEG signals and the discrepancies between the recorded data from different subjects in visual object recognition task make classification of cognitive states of subjects a serious challenge. In this research, we present a framework for evaluating machine learning and wrapper channel selection algorithms used for classifying single-trial EEG signals recorded in response to photographic stimuli. It is shown that by correctly mapping the entire EEG data space to informative feature spaces (IFS), the performance of the classification methods can improve significantly. Results outperform the state-of-the-art results and confirm efficiency of the proposed feature selection methods in capturing the most informative EEG channels. This can help to achieve high separability of object categories in single-trial visual object recognition task.

EEG-Single-Channel Envelope Synchronisation and Classification for Seizure Detection and Prediction

This paper tackles the complex issue of detecting and classifying epileptic seizures whilst maintaining the total calculations at a minimum. Where many systems depend on the coupling between multiple sources, leading to hundreds of combinations of electrodes, our method calculates the instantaneous phase between non-identical upper and lower envelopes of a single-electroencephalography channel reducing the workload to the total number of electrode points. From over 600 h of simulations, our method shows a sensitivity and specificity of 100% for high false-positive rates and 83% and 75%, respectively, for moderate to low false positive rates, which compares well to both single- and multi-channel-based methods. Furthermore, pre-ictal variations in synchronisation were detected in over 90% of patients implying a possible prediction system.

Roughness-Length-Based Characteristic Analysis of Intracranial EEG and Epileptic Seizure Prediction

To identify precursors of epileptic seizures, an EEG characteristic analysis is carried out based on a roughness-length method, where fractal dimensions and intercept values are extracted to measure the structure complexity and the amplitude roughness of EEG signals in different phases. Using the significant changes of the fractal dimension and intercept in the preictal phase with respect to those in the interictal phase, a patient-specific seizure prediction algorithm is then proposed by combining with a gradient boosting classifier. The probabilistic outputs of the trained gradient boosting classifier are further processed by threshold comparison and rule-based judgment to distinguish preictal EEG from interictal EEG and to generate seizure alerts. The prediction algorithm was evaluated on 20 patients’ intracranial EEG recordings from the Freiburg EEG database, which contains the preictal periods of 65 seizures and 499[Formula: see text]h interictal EEG. Setting the seizure prediction horizon as 2[Formula: see text]min, averaged sensitivity values of 90.42% and 91.67% with averaged false prediction rates of 0.12/h and 0.10/h were achieved for seizure occurrence periods of 30 and 50[Formula: see text]min, respectively. These results demonstrate the ability of fractal dimension and intercept metrics in predicting the occurrence of seizures.

[Application of scalp electroencephalogram in treatment of refractory epilepsy with vagus nerve stimulation]

Electroencephalogram (EEG) has been an important tool for scientists to study epilepsy and evaluate the treatment of epilepsy for half a century, since epilepsy seizures are caused by the diffusion of excessive discharge of brain neurons. This paper reviews the clinical application of scalp EEG in the treatment of intractable epilepsy with vagus nerve stimulation (VNS) in the past 30 years. It mainly introduces the prediction of the therapeutic effect of VNS on intractable epilepsy based on EEG characteristics and the effect of VNS on EEG of patients with intractable epilepsy, and expounds some therapeutic mechanisms of VNS. For predicting the efficacy of VNS based on EEG characteristics, EEG characteristics such as epileptiform discharge, polarity of slow cortical potential changes, changes of EEG symmetry level and changes of EEG power spectrum are described. In view of the influence of VNS treatment on patients' EEG characteristics, the change of epileptiform discharge, power spectrum, synchrony, brain network and amplitude of event-related potential P300 are described. Although no representative EEG markers have been identified for clinical promotion, this review paves the way for prospective studies of larger patient populations in the future to better apply EEG to the clinical treatment of VNS, and provides ideas for predicting VNS efficacy, assessing VNS efficacy, and understanding VNS treatment mechanisms, with broad medical and scientific implications.

Scalp electroencephalography (sEEG) based advanced prediction of epileptic seizure time and identification of epileptogenic region

Objectives Epilepsy is characterized by uncontrollable seizure during which consciousness of patient is disturbed. Prediction of the seizure in advance will increase the remedial possibilities for the patients suffering from epilepsy. An automated system for seizure prediction is important for seizure enactment, prevention of sudden unexpected deaths and to avoid seizure related injuries. Methods This paper proposes the prediction of an upcoming seizure by analyzing the 23 channel non-stationary EEG signal. EEG signal is divided into smaller segments to change it into quasi-stationary data using an overlapping moving window. Brain region is marked into four regions namely left hemisphere, right hemisphere, central region and temporal region to identify the epileptogenic region. The epileptogenic region shows significant variations during pre-ictal state in comparison to the other regions. So, seizure prediction is carried out by analyzing EEG signals from this region. Seizure prediction is proposed using features extracted from both time and frequency domain. Relative entropy and relative energy are extracted from wavelet transform and Pearson correlation coefficient is obtained from time domain EEG signal. Extracted features have been smoothened using moving average filter. First order derivative of relative features have been used to normalize the intervariability before deciding the threshold for marking the prediction of seizure. Results Isolated seizures where pre-ictal duration of more than 1 h is reported has been detected with an accuracy of 92.18% with precursory warning 18 min in advance and seizure confirmation 12 min in advance. An overall accuracy of 83.33% with false positive alarm rate of 0.01/h has been obtained for all seizure cases with average prediction time of 9.9 min.

Automated classification of five seizure onset patterns from intracranial electroencephalogram signals

OBJECTIVE

The electroencephalographic (EEG) signals contain information about seizures and their onset location. There are several seizure onset patterns reported in the literature, and these patterns have clinical significance. In this work, we propose a system to automatically classify five seizure onset patterns from intracerebral EEG signals.

METHODS

The EEG was segmented by clinicians indicating the start and end time of each seizure onset pattern, the channels involved at onset and the seizure onset pattern. Twelve features that represent the time domain characteristics and signal complexity were extracted from 663 seizures channels of 24 patients. The features were used for classification of the patterns with support vector machine - Error-Correcting Output Codes (SVM-ECOC). Three patient groups with a similar number of seizure segments were created, and one group was used for testing and the rest for training. This test was repeated by rotating the testing and training data.

RESULTS

The feature space formed by both time domain and multiscale sample entropy features perform well in classification of the data. An overall accuracy of 80.7% was obtained with these features and a linear kernel of SVM-ECOC.

CONCLUSIONS

The seizure onset patterns consist of varied time and complexity characteristics. It is possible to automatically classify various seizure onset patterns very similarly to visual classification.

SIGNIFICANCE

The proposed system could aid the medical team in assessing intracerebral EEG by providing an objective classification of seizure onset patterns.

Prediction error connectivity: A new method for EEG state analysis

Several models have been proposed to explain brain regional and interregional communication, the majority of them using methods that tap the frequency domain, like spectral coherence. Considering brain interareal communication as binary interactions, we describe a novel method devised to predict dynamics and thus highlight abrupt changes marked by unpredictability. Based on a variable-order Markov model algorithm developed in-house for data compression, the prediction error connectivity (PEC) estimates network transitions by calculating error matrices (EMs). We analysed 20 h of EEG signals of virtual networks generated with a neural mass model. Subnetworks changed through time (2 of 5 signals), from normal to normal or pathological states. PEC was superior to spectral coherence in detecting all considered transitions, especially in broad and ripple bands. Subsequently, EMs of real data were classified using a support vector machine in order to capture the transition from interictal to preictal state and calculate seizure risk. A single seizure was randomly selected for training. Through this approach it was possible to establish a threshold that the calculated risk consistently overcame minutes before the events. Using either spectral coherence or PEC we created 1000 models that successfully predicted 6 seizures (100% sensibility), a whole cluster recorded in a patient with hippocampal epilepsy. However, PEC resulted superior to coherence in terms of true seizure free time and amount of false warnings. Indeed, the best PEC model predicted 96% of interictal time (vs. 83% of coherence) of about 20 h of stereo-EEG. This analysis was extended to patients with neo/mesocortical temporal, neocortical frontal, parietal and occipital lobe epilepsy. Again PEC showed high performance, allowing the prediction of 31 events distributed across 10 days with ROC AUCs that reached 98% (average 93 ± 5%) in 6 different patients. Moreover, considering another state transition, PEC could classify and forecast up to 88% (average 85 ± 3%) of the REM phase both in deep and scalp EEG. In conclusion, PEC is a novel approach that relies on pattern analysis in the time-domain. We believe that this method can be successfully employed both for the study of brain connectivity, and also implemented in real-life solutions for seizure detection and prediction.

Automated seizure prediction

In the past two decades, significant advances have been made on automated electroencephalogram (EEG)-based diagnosis of epilepsy and seizure detection. A number of innovative algorithms have been introduced that can aid in epilepsy diagnosis with a high degree of accuracy. In recent years, the frontiers of computational epilepsy research have moved to seizure prediction, a more challenging problem. While antiepileptic medication can result in complete seizure freedom in many patients with epilepsy, up to one-third of patients living with epilepsy will have medically intractable epilepsy, where medications reduce seizure frequency but do not completely control seizures. If a seizure can be predicted prior to its clinical manifestation, then there is potential for abortive treatment to be given, either self-administered or via an implanted device administering medication or electrical stimulation. This will have a far-reaching impact on the treatment of epilepsy and patient's quality of life. This paper presents a state-of-the-art review of recent efforts and journal articles on seizure prediction. The technologies developed for epilepsy diagnosis and seizure detection are being adapted and extended for seizure prediction. The paper ends with some novel ideas for seizure prediction using the increasingly ubiquitous machine learning technology, particularly deep neural network machine learning.

Estimating The Quality of Electroconvulsive Therapy Induced Seizures Using Decision Tree and Fuzzy Inference System Classifiers

Electroconvulsive therapy (ECT) is an effective and widely used treatment for major depressive disorder, in which a brief electric current is passed through the brain to trigger a brief seizure. This study aims to identify seizure quality rating by utilizing a set of seizure parameters. We used 750 ECT EEG recordings in this experiment. Four seizure related parameters, (time of slowing, regularity, stereotypy and post-ictal suppression) are used as inputs to two classifiers, decision tree and fuzzy inference system (FIS), to predict seizure quality ratings. The two classifiers produced encouraging results with error rate of 0.31 and 0.25 for FIS and decision tree, respectively. The classification results show that the four seizure parameters provide relevant information about the rating of seizure quality. Automatic scoring of seizure quality may be beneficial to clinicians working in this field.

Towards an Online Seizure Advisory System-An Adaptive Seizure Prediction Framework Using Active Learning Heuristics

In the last decade, seizure prediction systems have gained a lot of attention because of their enormous potential to largely improve the quality-of-life of the epileptic patients. The accuracy of the prediction algorithms to detect seizure in real-world applications is largely limited because the brain signals are inherently uncertain and affected by various factors, such as environment, age, drug intake, etc., in addition to the internal artefacts that occur during the process of recording the brain signals. To deal with such ambiguity, researchers transitionally use active learning, which selects the ambiguous data to be annotated by an expert and updates the classification model dynamically. However, selecting the particular data from a pool of large ambiguous datasets to be labelled by an expert is still a challenging problem. In this paper, we propose an active learning-based prediction framework that aims to improve the accuracy of the prediction with a minimum number of labelled data. The core technique of our framework is employing the Bernoulli-Gaussian Mixture model (BGMM) to determine the feature samples that have the most ambiguity to be annotated by an expert. By doing so, our approach facilitates expert intervention as well as increasing medical reliability. We evaluate seven different classifiers in terms of the classification time and memory required. An active learning framework built on top of the best performing classifier is evaluated in terms of required annotation effort to achieve a high level of prediction accuracy. The results show that our approach can achieve the same accuracy as a Support Vector Machine (SVM) classifier using only 20 % of the labelled data and also improve the prediction accuracy even under the noisy condition.

Epileptic Seizure Prediction Using Diffusion Distance and Bayesian Linear Discriminate Analysis on Intracranial EEG

Epilepsy is a chronic neurological disorder characterized by sudden and apparently unpredictable seizures. A system capable of forecasting the occurrence of seizures is crucial and could open new therapeutic possibilities for human health. This paper addresses an algorithm for seizure prediction using a novel feature — diffusion distance (DD) in intracranial Electroencephalograph (iEEG) recordings. Wavelet decomposition is conducted on segmented electroencephalograph (EEG) epochs and subband signals at scales 3, 4 and 5 are utilized to extract the diffusion distance. The features of all channels composing a feature vector are then fed into a Bayesian Linear Discriminant Analysis (BLDA) classifier. Finally, postprocessing procedure is applied to reduce false prediction alarms. The prediction method is evaluated on the public intracranial EEG dataset, which consists of 577.67[Formula: see text]h of intracranial EEG recordings from 21 patients with 87 seizures. We achieved a sensitivity of 85.11% for a seizure occurrence period of 30[Formula: see text]min and a sensitivity of 93.62% for a seizure occurrence period of 50[Formula: see text]min, both with the seizure prediction horizon of 10[Formula: see text]s. Our false prediction rate was 0.08/h. The proposed method yields a high sensitivity as well as a low false prediction rate, which demonstrates its potential for real-time prediction of seizures.

Epileptic Seizure Prediction Using CSP and LDA for Scalp EEG Signals

This paper presents a patient-specific epileptic seizure predication method relying on the common spatial pattern- (CSP-) based feature extraction of scalp electroencephalogram (sEEG) signals. Multichannel EEG signals are traced and segmented into overlapping segments for both preictal and interictal intervals. The features extracted using CSP are used for training a linear discriminant analysis classifier, which is then employed in the testing phase. A leave-one-out cross-validation strategy is adopted in the experiments. The experimental results for seizure prediction obtained from the records of 24 patients from the CHB-MIT database reveal that the proposed predictor can achieve an average sensitivity of 0.89, an average false prediction rate of 0.39, and an average prediction time of 68.71 minutes using a 120-minute prediction horizon.

Prediction of epileptic seizures from EEG using analysis of ictal rules on Poincaré plane

BACKGROUND AND OBJECTIVE

Epilepsy is a neurological disorder that causes recurrent and abrupt seizures which makes the patients insecure. Predicting seizures can reduce the burdens of this disorder.

METHODS

A new approach in seizure prediction is presented that includes a novel technique in feature extraction from EEG. The algorithm firsts creates an embedding space from EEG time series. Then it takes samples with most of the information using an optimized and data specific Poincare plane. In order to quantify small dynamics on the Poincare plane, based on the order of locations of Poincaré samples in the sequence, 64 fuzzy rules in each channel are defined. Features are extracted based on the frequency distribution of these fuzzy rules in each minute. Then features with higher variance are selected as ictal features and again reduced using PCA. Finally, in order to evaluate how these innovative features can increase the performance of the seizure prediction algorithm, the transition from interictal to preictal state is scored utilizing SVM.

RESULTS

The algorithm is tested on 460 h of EEG from 19 patients of Freiburg dataset who had at least 3 seizures. Considering maximum Seizure Prediction Horizon of 42 minutes, average sensitivity was 91.8 - 96.6% and average false prediction rate was 0.05 - 0.08/h.

CONCLUSIONS

The presented algorithm shows a better performance and more robustness compare to most of existing methods, and shows power in extracting optimal features from EEG.

Seizure prediction in patients with focal hippocampal epilepsy

OBJECTIVE

We evaluated the performance of our previously developed seizure prediction approach on thirty eight seizures from ten patients with focal hippocampal epilepsy.

METHODS

The seizure prediction system was developed based on the extraction of correlation dimension, correlation entropy, noise level, Lempel-Ziv complexity, largest Lyapunov exponent, and nonlinear interdependence from segments of intracranial EEG.

RESULTS

Our results showed an average sensitivity of 86.7% and 92.9%, an average false prediction rate of 0.126 and 0.096/h, and an average minimum prediction time of 14.3 and 33.3min, respectively, using seizure occurrence periods of 30 and 50min and a seizure prediction horizon of 10s. Two-third of the analyzed seizures showed significantly increased complexity in periods prior to the seizures in comparison with baseline. In four patients, strong bidirectional connectivities between epileptic contacts and the surrounding areas were observed. However, in five patients, unidirectional functional connectivities in preictal periods were observed from remote areas to epileptogenic zones.

CONCLUSIONS

Overall, preictal periods in patients with focal hippocampal epilepsy were characterized with patient-specific changes in univariate and bivariate nonlinear measures.

SIGNIFICANCE

The spatio-temporal characterization of preictal periods may help to better understand the mechanism underlying seizure generation in patients with focal hippocampal epilepsy.

An efficient automatic workload estimation method based on electrodermal activity using pattern classifier combinations

Automatic workload estimation has received much attention because of its application in error prevention, diagnosis, and treatment of neural system impairment. The development of a simple but reliable method using minimum number of psychophysiological signals is a challenge in automatic workload estimation. To address this challenge, this paper presented three different decomposition techniques (Fourier, cepstrum, and wavelet transforms) to analyze electrodermal activity (EDA). The efficiency of various statistical and entropic features was investigated and compared. To recognize different levels of an arithmetic task, the features were processed by principal component analysis and machine-learning techniques. These methods have been incorporated into a workload estimation system based on two types: feature-level and decision-level combinations. The results indicated the reliability of the method for automatic and real-time inference of psychological states. This method provided a quantitative estimation of the workload levels and a bias-free evaluation approach. The high-average accuracy of 90% and cost effective requirement were the two important attributes of the proposed workload estimation system. New entropic features were proved to be more sensitive measures for quantifying time and frequency changes in EDA. The effectiveness of these measures was also compared with conventional tonic EDA measures, demonstrating the superiority of the proposed method in achieving accurate estimation of workload levels.

EEG analysis via multiscale Lempel-Ziv complexity for seizure detection

Robust seizure detection and seizure prediction continues to be a challenge. Lempel-Ziv Complexity (LZC) is one of the features that has shown to be relevant in seizure detection. Recent work has shown that augmenting LZC can be beneficial to emphasize variations in amplitude or frequency when analyzing biomedical signals. In this paper, we present a first look into evaluating the feasibility of using a recently proposed feature stemmed from LZC, namely the Multiscale Lempel-Ziv Complexity (MLZC) for seizure detection. MLZC does not allow the high-frequency signal components to be overwhelmed by the low frequency signal components when calculating complexity values. We compare MLZC and LZC for identifying seizures for three cases and show MLZC can provide a clear separation between non-ictal and ictal periods for all three cases using a single threshold over 7 recordings and 7 seizures per patient, whereas LZC provided such a clear separation for only one of the patients.

Real-time seizure prediction using RLS filtering and interpolated histogram feature based on hybrid optimization algorithm of Bayesian classifier and Hunting search

BACKGROUND AND OBJECTIVES

Epileptic seizure prediction using EEG signal analysis is an important application for drug therapy and pediatric patient monitoring. Time series estimation to obtain the future samples of EEG signal has vital role for detecting seizure attack. In this paper, a novel density-based real-time seizure prediction algorithm based on a trained offline seizure detection algorithm is proposed.

METHODS

In the offline seizure detection procedure, after signal preprocessing, histogram-based statistical features are extracted from signal probability distribution. By defining a deterministic polynomial model on the normalized histogram, a novel syntactic feature that is named Interpolated Histogram Feature (IHF) is proposed. Moreover, with this feature, Seizure Distribution Model (SDM) as a descriptor of the seizure and non-seizure signals is presented. By using a novel hybrid optimization algorithm based on Bayesian classifier and Hunting Search (HuS) algorithm, the optimal features are selected. To detect the seizure attacks in the online mode, a Multi-Layer Perceptron (MLP) classifier is trained with the optimal features in the offline procedure. For online prediction, the enhanced Recursive Least Square (RLS) filter is applied to estimate sample-by-sample of the EEG signal. Also, a density-based signal tracking scenario is introduced to update and tune the parameters of RLS filtering algorithm.

RESULTS

Our prediction algorithm is evaluated on 104 hours of EEG signals recorded from 23 pediatric patients. Our online signal prediction algorithm provides the accuracy rate of 86.56% and precision rate of 86.53% simultaneously using the trained MLP classifier from the offline mode. The recall rate of seizure prediction is 97.27% and the false prediction rate of 0.00215 per hour is achieved as well. Ultimately, the future samples of EEG signal are estimated, and the time of seizure signal prediction is also converged to 6.64 seconds.

CONCLUSION

In our proposed real-time algorithm, by implementing a density-based signal tracking scenario, the future samples of signal with suitable time is predicted and the seizure is detected based on the optimal features from the IHF and histogram-based statistical features with acceptable performance.

All together now: Analogies between chimera state collapses and epileptic seizures

Conceptually and structurally simple mathematical models of coupled oscillator networks can show a rich variety of complex dynamics, providing fundamental insights into many real-world phenomena. A recent and not yet fully understood example is the collapse of coexisting synchronous and asynchronous oscillations into a globally synchronous motion found in networks of identical oscillators. Here we show that this sudden collapse is promoted by a further decrease of synchronization, rather than by critically high synchronization. This strikingly counterintuitive mechanism can be found also in nature, as we demonstrate on epileptic seizures in humans. Analyzing spatiotemporal correlation profiles derived from intracranial electroencephalographic recordings (EEG) of seizures in epilepsy patients, we found a pronounced decrease of correlation at the seizure onsets. Applying our findings in a closed-loop control scheme to models of coupled oscillators in chimera states, we succeed in both provoking and preventing outbreaks of global synchronization. Our findings not only advance the understanding of networks of coupled dynamics but can open new ways to control them, thus offering a vast range of potential new applications.

Assessing directionality and strength of coupling through symbolic analysis: an application to epilepsy patients

Inferring strength and direction of interactions from electroencephalographic (EEG) recordings is of crucial importance to improve our understanding of dynamical interdependencies underlying various physiological and pathophysiological conditions in the human epileptic brain. We here use approaches from symbolic analysis to investigate--in a time-resolved manner--weighted and directed, short- to long-ranged interactions between various brain regions constituting the epileptic network. Our observations point to complex spatial-temporal interdependencies underlying the epileptic process and their role in the generation of epileptic seizures, despite the massive reduction of the complex information content of multi-day, multi-channel EEG recordings through symbolization. We discuss limitations and potential future improvements of this approach.

Automatic epileptic seizure detection using scalp EEG and advanced artificial intelligence techniques

The epilepsies are a heterogeneous group of neurological disorders and syndromes characterised by recurrent, involuntary, paroxysmal seizure activity, which is often associated with a clinicoelectrical correlate on the electroencephalogram. The diagnosis of epilepsy is usually made by a neurologist but can be difficult to be made in the early stages. Supporting paraclinical evidence obtained from magnetic resonance imaging and electroencephalography may enable clinicians to make a diagnosis of epilepsy and investigate treatment earlier. However, electroencephalogram capture and interpretation are time consuming and can be expensive due to the need for trained specialists to perform the interpretation. Automated detection of correlates of seizure activity may be a solution. In this paper, we present a supervised machine learning approach that classifies seizure and nonseizure records using an open dataset containing 342 records. Our results show an improvement on existing studies by as much as 10% in most cases with a sensitivity of 93%, specificity of 94%, and area under the curve of 98% with a 6% global error using a k-class nearest neighbour classifier. We propose that such an approach could have clinical applications in the investigation of patients with suspected seizure disorders.

Approximate entropy and support vector machines for electroencephalogram signal classification

The automatic detection and identification of electroencephalogram waves play an important role in the prediction, diagnosis and treatment of epileptic seizures. In this study, a nonlinear dynamics index-approximate entropy and a support vector machine that has strong generalization ability were applied to classify electroencephalogram signals at epileptic interictal and ictal periods. Our aim was to verify whether approximate entropy waves can be effectively applied to the automatic real-time detection of epilepsy in the electroencephalogram, and to explore its generalization ability as a classifier trained using a nonlinear dynamics index. Four patients presenting with partial epileptic seizures were included in this study. They were all diagnosed with neocortex localized epilepsy and epileptic foci were clearly observed by electroencephalogram. The electroencephalogram data form the four involved patients were segmented and the characteristic values of each segment, that is, the approximate entropy, were extracted. The support vector machine classifier was constructed with the approximate entropy extracted from one epileptic case, and then electroencephalogram waves of the other three cases were classified, reaching a 93.33% accuracy rate. Our findings suggest that the use of approximate entropy allows the automatic real-time detection of electroencephalogram data in epileptic cases. The combination of approximate entropy and support vector machines shows good generalization ability for the classification of electroencephalogram signals for epilepsy.

Patient specific seizure prediction system using Hilbert spectrum and Bayesian networks classifiers

The aim of this paper is to develop an automated system for epileptic seizure prediction from intracranial EEG signals based on Hilbert-Huang transform (HHT) and Bayesian classifiers. Proposed system includes decomposition of the signals into intrinsic mode functions for obtaining features and use of Bayesian networks with correlation based feature selection for binary classification of preictal and interictal recordings. The system was trained and tested on Freiburg EEG database. 58 hours of preictal data, 40-minute data blocks prior to each of 87 seizures collected from 21 patients, and 503.1 hours of interictal data were examined resulting in 96.55% sensitivity with 0.21 false alarms per hour, 13.896% average proportion of time spent in warning, and 33.21 minutes of average detection latency using 30-second EEG segments with 50% overlap and a simple postprocessing technique resulting in a decision (a seizure is expected/not expected) every 5 minutes. High sensitivity and low false positive rate with reasonable detection latency show that HHT based features are acceptable for patient specific seizure prediction from intracranial EEG data. Time spent for testing an EEG segment was 4.1451 seconds on average, which makes the system viable for use in real-time seizure control systems.

A low computation cost method for seizure prediction

The dynamic changes of electroencephalograph (EEG) signals in the period prior to epileptic seizures play a major role in the seizure prediction. This paper proposes a low computation seizure prediction algorithm that combines a fractal dimension with a machine learning algorithm. The presented seizure prediction algorithm extracts the Higuchi fractal dimension (HFD) of EEG signals as features to classify the patient's preictal or interictal state with Bayesian linear discriminant analysis (BLDA) as a classifier. The outputs of BLDA are smoothed by a Kalman filter for reducing possible sporadic and isolated false alarms and then the final prediction results are produced using a thresholding procedure. The algorithm was evaluated on the intracranial EEG recordings of 21 patients in the Freiburg EEG database. For seizure occurrence period of 30 min and 50 min, our algorithm obtained an average sensitivity of 86.95% and 89.33%, an average false prediction rate of 0.20/h, and an average prediction time of 24.47 min and 39.39 min, respectively. The results confirm that the changes of HFD can serve as a precursor of ictal activities and be used for distinguishing between interictal and preictal epochs. Both HFD and BLDA classifier have a low computational complexity. All of these make the proposed algorithm suitable for real-time seizure prediction.