No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Nonlinear Dynamics of Electrocorticographic Data in Temporal Lobe Epilepsy
... /fneur. . (4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17), the first prospective seizure prediction algorithm running real-time on continuous EEG data was developed by Iasemidis et al. (18). Non-linear features of the EEG, such as the largest Lyapunov exponent and phase changes in the state space, were extracted over time to identify dynamical spatial entrainment changes in the preictal period between critical brain sites. ...
The recurrent and unpredictable nature of seizures can lead to unintentional injuries and even death. The rapid development of electroencephalogram (EEG) and Artificial Intelligence (AI) technologies has made it possible to predict seizures in real-time through brain-machine interfaces (BCI), allowing advanced intervention. To date, there is still much room for improvement in predictive seizure models constructed by EEG using machine learning (ML) and deep learning (DL). But, the most critical issue is how to improve the performance and generalization of the model, which involves some confusing conceptual and methodological issues. This review focuses on analyzing several factors affecting the performance of seizure prediction models, focusing on the aspects of post-processing, seizure occurrence period (SOP), seizure prediction horizon (SPH), and algorithms. Furthermore, this study presents some new directions and suggestions for building high-performance prediction models in the future. We aimed to clarify the concept for future research in related fields and improve the performance of prediction models to provide a theoretical basis for future applications of wearable seizure detection devices.
... Iasemidis and Sackellares were the first group to apply nonlinear dynamical techniques, particularly methods based upon the principal Lyapunov exponent (PLE), for predicting seizures beginning in the late 1980s. This group has demonstrated evidence of seizure precursors in a variety of data sets, ranging from one to multiple channels and epochs spanning minutes to hours [23]–[27]. In their research, seizure onset prediction was reported from one to sixty minutes prior to electrographic seizure onset in some data sets. ...
Epileptic seizure prediction has steadily evolved from its conception in the 1970s, to proof-of-principle experiments in the late 1980s and 1990s, to its current place as an area of vigorous, clinical and laboratory investigation. As a step toward practical implementation of this technology in humans, we present an individualized method for selecting electroencephalogram (EEG) features and electrode locations for seizure prediction focused on precursors that occur within ten minutes of electrographic seizure onset. This method applies an intelligent genetic search process to EEG signals simultaneously collected from multiple intracranial electrode contacts and multiple quantitative features derived from these signals. The algorithm is trained on a series of baseline and preseizure records and then validated on other, previously unseen data using split sample validation techniques. The performance of this method is demonstrated on multiday recordings obtained from four patients implanted with intracranial electrodes during evaluation for epilepsy surgery. An average probability of prediction (or block sensitivity) of 62.5% was achieved in this group, with an average block false positive (FP) rate of 0.2775 FP predictions/h, corresponding to 90.47% specificity. These findings are presented as an example of a method for training, testing and validating a seizure prediction system on data from individual patients. Given the heterogeneity of epilepsy, it is likely that methods of this type will be required to configure intelligent devices for treating epilepsy to each individual's neurophysiology prior to clinical deployment.
The modeling of dynamic behavior of systems is a ubiquitous problem in all facets of human endeavors. Importantly so, dynamical systems have been studied and modeled since the nineteenth century and currently applied in almost all branches of sciences and engineering including social sciences. The development of computers and scientific/numerical methods has accelerated the pace of new developments in modeling both linear and nonlinear dynamical systems. However, modeling complex physical system behaviors as nonlinear dynamical systems is still difficult and challenging. General approaches to solving such systems typically fail and require personalized problem dependent techniques to satisfy the constraints imposed based on the initial conditions to predict state space trajectories. In addition, they require enormous computational power available on supercomputers. Numerical tools such as HPCmatlab enable rapid prototyping of algorithms for large scale computations and data analysis. BigData applications are computationally intensive and I/O bound. An example, state of the art case study involving big data of epileptic seizure prediction and control is presented. The nonlinear dynamical model is based on the biology of the brain and its neurons, chaotic systems, nonlinear signal processing, and feedback and adaptive systems. The goal is to develop new feedback controllers for the suppression of epileptic seizures based on electroencephalographic (EEG) data by altering the brain dynamics through the use of electrical stimulation. The research is expected to contribute to new modes of treatment for epilepsy and other dynamical brain disorders.
We propose a novel approach for detecting precursors to epileptic seizures in intracranial electroencephalograms (iEEGs), which is based on the analysis of system dynamics. In the proposed scheme, the largest Lyapunov exponent (LLE) of wavelet entropy of the segmented EEG signals are considered as the discriminating features. Such features are processed by a support vector machine classifier, whose outcomes (the label and its probability for each LLE) are post-processed and fed into a novel decision function to determine whether the corresponding segment of the EEG signal contains a precursor to an epileptic seizure. The proposed scheme is applied to the Freiburg data set, and the results show that seizure precursors are detected in a time frame that unlike other existing schemes is very much convenient to patients, with the sensitivity of 100% and negligible false positive detection rates.
Epileptic seizures are abnormal, excessive, synchronous and unprovoked neural activity occuring in the brain affecting 2% of the world population. Epileptic patients mostly generates interictal EEG which on accurate detection provides vital information for epileptic seizure prediction. Nonconvulsive epilepsy, most commonly occuring epilepsy, results in continuous seizing of brain and the patient becomes unresponsive during the episode. This type of epileptic seizures show minimal symptoms and are often under-diagnosed resulting in late treatment. A multi level classification (epileptic, interictal and normal) providing interictal information in addition to normal and epileptic informations proves to be highly useful to Neurologists in the prediction and diagnosis of nonconvulsive epileptic seizures. An automated multilevel SVM based nonconvulsive epileptic seizure detection method utilizing a single feature vector (enabling fast classification) calculated from the fifth detail wavelet coefficients of the given EEG data is proposed here. Case study on different features like energy, mean energy, entropy, mean cross correlation, mean curve length, coefficient of variation, interquartile range and median absolute deviation yielded an optimum single feature. Energy, entropy, interquartile range and median absolute deviation were effective (above 95% accuracy) for the multi-level classification of EEG data with interquartile range being optimum feature with 99.96% accuracy and very low runtime of less than six seconds.
We discuss the key features of theoretical models of networks of coupled chaotic oscillators that produce seizure-like events and bear striking similarities to dynamics of epileptic seizures. Our long-term objective is to understand the basic functional mechanisms that can produce seizures and may ultimately lead to strategies for seizure suppression and control. We show that, from a dynamical systems point of view, a plausible cause of seizures is a pathological feedback in the brain circuitry. This suggests new seizure control approaches, as well as systematic methods to tune existing ones. While the suggested models and control approaches are far from being considered optimal for epileptic seizures, they have interesting physical interpretation and implications for treatments of epilepsy. They also have close ties with a variety of recent practical observations in the human and animal epileptic brain, and with theories from adaptive systems, optimization and chaos.
The presence of an on-line seizure detection system could drive an antiepileptic stimulator in real time to suppress seizure generation and to enhance the patients' safety and quality of life. In this paper, the continuous long-term EEGs of three Wistar rats with spontaneous temporal lobe seizure were analyzed. We proposed the development of an energy efficient real-time seizure detection method that employs a hierarchical architecture. The first stage was used to fast detect the seizure-like EEG segment, and a classifier was utilized in the second stage for final confirmation. Only when a suspected seizure segment is found, the second stage is activated. With 2-staged architecture, it saved about 99.4% computation energy in the experiment. Therefore, it is useful to improve the longevity of the closed-loop seizure control system. Three classifiers, linear discriminant analysis (LDA), quadratic discriminant analysis (QDA) and support vector machine (SVM), were applied for comparison. From the experimental results, three classifiers yielded the comparable performances. However, considering of the trade-off between detection performances and power consumption, LDA which yielded the 100% detection rate, 0.22 FP/hr, and 1.69 s detection latency is suggested for a portable closed-loop seizure controller.
Epilepsy is one of the most common neurological disorders, characterized by recurrent seizures. Being able to predict impending seizures could greatly improve the lives of patients with epilepsy. In this study, we propose a new adaptive learning approach for online seizure prediction based on analysis of electroencephalogram (EEG) recordings. For each individual patient, we construct baseline patterns of normal and preseizure EEG samples, continuously monitor sliding windows of EEG recordings, and classify each window to normal or preseizure using a $(K)$-nearest-neighbor (KNN) method. A new reinforcement learning algorithm is proposed to continuously update both normal and preseizure baseline patterns based on the feedback from prediction result of each window. The proposed approach was evaluated on EEG data from 10 patients with epilepsy. For each one of the 10 patients, the adaptive approach was trained using the recordings containing the first half of seizure occurrences, and tested prospectively on the subsequent recordings. Using a 150-minute prediction horizon, our approach achieved 73 percent sensitivity and 67 percent specificity on average over 10 patients. This result is shown to be far better than those of a nonupdate prediction scheme and two native prediction schemes.
Previous work has demonstrated that some dynamic properties of intracranial EEG signals are indicative of epileptic seizures and hence could be used for prediction in order to realize counter measures. However, most previous studies only investigated predictability via offline analysis of EEG signals as compared to actually predicting seizures in a setting applicable to implantable devices. Here we address this problem, which calls for simple and fast online methods, and based on previous offline analyses we hypothesize that prediction can be further improved when using multiple features to detect the preictal patterns. We propose a simple adaptive online method (an evolving neuro-fuzzy model) to adaptively learn such combined features. The classifier starts out with a simple structure and patient-independent parameters and then grows into a personal seizure predictor as recursive methods tune the model structure and parameters. We apply the adaptive classifier to a publicly available database of intracranial recordings from 21 patients and demonstrate that seizure prediction is improved with our online method as compared to offline non-adaptive techniques. We show that our method is robust with respect to those few model parameters, which are not adapted. Moreover, as we report the performance on data from a publicly available seizure database, our results can serve as a yardstick for future method developments.
Recently, interest has turned to the mathematical concept of chaos as an explanation for a variety of complex processes in nature. Chaotic systems, among other characteristics, can produce what appears to be random output. Another property of chaotic systems is that they may exhibit abrupt intermittent transitions between highly ordered and disordered states. Because of this property, it is hypothesized that epilepsy may be an example of chaos. In this review, some of the basic concepts of nonlinear dynamics and chaos are illustrated. Mathematical techniques developed to study the properties of nonlinear dynamical systems are outlined. Finally, the results of applying these techniques to the study of human epilepsy are discussed. The application of these powerful and novel mathematical techniques to analysis of the electroencephalogram has provided new insights into the epileptogenic process and may have considerable utility in the diagnosis and treatment of epilepsy.
It is critical to set up a precise and feasible monitoring system for a variety of animal and human studies. A multichannel wireless system for monitoring physiological signals of freely moving rats is presented. This system combines electroencephalogram (EEG) and acceleration signals, enabling the study of association between brain and behavior. A combination of EEG and accelerometers eliminates the necessity for complicated video installation as well as time-consuming and tedious analysis of recorded videos. The IEEE 802.15.4 based wireless communication frees the experimental subject from the hassle of wires and reduces wire artifacts during recording. Long-period continuous recording was possible because of the low power feature of the system. Methods for automatic wake–sleep state discrimination and temporal lobe epileptic seizure detection are also proposed to demonstrate the advantages of the system. An accuracy of up to 96.22% for the automatic discrimination of wake–sleep states is an advantage of our system. In addition, the detection of amygdala-kindling temporal lobe seizures reaches 100% with zero false alarms, greatly saving manpower in the identification of temporal lobe epilepsy.
Epilepsy is a major neurological disorder characterized by intermittent paroxysmal neuronal electrical activity, that may
remain localized or spread, and severely disrupt the brain’s normal operation. Epileptic seizures are typical manifestations
of such pathology. It is in the last 20 years that prediction and control of epileptic seizures has been the subject of intensive
interdisciplinary research. In this communication, we investigate epilepsy from the point of view of pathology of the dynamics
of the electrical activity of the brain. In this framework, we revisit two critical aspects of the dynamics of epileptic seizures
– the seizure predictability and seizure resetting – that may prove to be the keys for improved seizure prediction and seizure
control schemes. We use human EEG data and the concepts of spatial synchronization of chaos, phase and energy to first show
that seizures could be predictable in the order of tens of minutes prior to their onset. We then present additional statistical
evidence that the pathology of the brain dynamics prior to seizures is reset mostly upon seizures’ occurrence, a phenomenon
we have called seizure resetting. Finally, using a biologically-plausible neural population mathematical model that can exhibit
seizure-like behavior, we provide evidence for the effectiveness of a recently devised seizure control scheme we have called
“feedback decoupling”. This scheme also provides an interesting dynamical model for ictogenesis (generation of seizures).
To understand basic functional mechanisms that cause epileptic seizures, the paper discusses some key features of theoretical
brain functioning models. The hypothesis is put forward that a plausible reason for seizures is pathological feedback in brain
circuitry. The analysis of such circuitry has an interesting physical interpretation and may be used to cure epilepsy.
In an effort to understand basic functional mechanisms that can produce epileptic seizures, and strategies for seizure suppression and control, we discuss some key features of theoretical models of networks of coupled chaotic oscillators that produce seizure-like events and bear striking similarities to dynamics of epileptic seizures. We show that a plausible cause of seizures is a pathological feedback in the brain circuitry. These results have interesting physical interpretation and implications for treatment of epilepsy. They also have close ties with a variety of recent practical observations in the human and animal epileptic brain, and with theories from adaptive systems, optimization, and chaos.
There is growing evidence that temporal lobe seizures are preceded by a preictal transition, characterized by a gradual dynamical change from asymptomatic interictal state to seizure. We herein report the first prospective analysis of the online automated algorithm for detecting the preictal transition in ongoing EEG signals. Such, the algorithm constitutes a seizure warning system. The algorithm estimates STL
max
, a measure of the order or disorder of the signal, of EEG signals recorded from individual electrode sites. The optimization techniques were employed to select critical brain electrode sites that exhibit the preictal transition for the warning of epileptic seizures. Specifically, a quadratically constrained quadratic 0-1 programming problem is formulated to identify critical electrode sites. The automated seizure warning algorithm was tested in continuous, long-term EEG recordings obtained from 5 patients with temporal lobe epilepsy. For individual patient, we use the first half of seizures to train the parameter settings, which is evaluated by ROC (Receiver Operating Characteristic) curve analysis. With the best parameter setting, the algorithm applied to all cases predicted an average of 91.7% of seizures with an average false prediction rate of 0.196 per hour. These results indicate that it may be possible to develop automated seizure warning devices for diagnostic and therapeutic purposes.
Approximately 1% of the world's population has epilepsy, and 25% of epilepsy patients cannot be treated sufficiently by any available therapy. If an automatic seizure-detection system was available, it could reduce the time required by a neurologist to perform an off-line diagnosis by reviewing electroencephalogram (EEG) data. It could produce an on-line warning signal to alert healthcare professionals or to drive a treatment device such as an electrical stimulator to enhance the patient's safety and quality of life. This paper describes a systematic evaluation of current approaches to seizure detection in the literature. This evaluation was then used to suggest a reliable, practical epilepsy detection method. The combination of complexity analysis and spectrum analysis on an EEG can perform robust evaluations on the collected data. Principle component analysis (PCA) and genetic algorithms (GAs) were applied to various linear and nonlinear methods. The best linear models resulted from using all of the features without other processing. For the nonlinear models, applying PCA for feature reduction provided better results than applying GAs. The feasibility of executing the proposed methods on a personal computer for on-line processing was also demonstrated.
In this paper, the Poincare map function as a one-dimensional first-return map is obtained by approximating the scatter plots of inter-peak interval (IPI) during preictal and postictal periods from invasive EEG recordings of nine patients suffering from medically intractable focal epilepsy. Evolutionary Algorithm (EA) is utilized for parameter estimation of the Poincare map. Bifurcation analyses of the iterated map reveal that as the neuronal activity progresses from preictal state toward the ictal event, the parameter values of the Poincare map move toward the bifurcation points. However, following the seizure occurrence and in the postictal period, these parameter values move away from the bifurcation points. Both flip and fold bifurcations are analyzed and it is demonstrated that in some cases the flip bifurcation and in other cases the fold bifurcation are the dynamical regime underlying epileptiform events. This information can offer insights into the dynamical nature and variability of the brain signals and consequently could help to predict and control seizure events.
To diagnose the structural disorders of brain, electroencephalography (EEG) is routinely used for observing the epileptic seizures in neurology clinics, which is one of the major brain disorders till today. In this work, we present a new, EEG- based, brain-state identification method which could form the basis for detecting epileptic seizure. We aim to classify the EEG signals and diagnose the epileptic seizures directly by using weighted locally linear embedding (WLLE) and support vector machine (SVM). Firstly, we use WLLE to do feature extraction of the EEG signal to obtain more compact representations of the internal characteristic and structure in the original data, which captures the information necessary for further manipulations. Then, SVM classifier is used to identify the seizures onset state from normal state of the patients.
Epilepsy is characterized by intermittent, paroxysmal, hypersynchronous electrical activity that may remain localized and/or spread and severely disrupt the brain's normal multitask and multiprocessing function. Epileptic seizures are the hallmarks of such activity. The ability to issue warnings in real time of impending seizures may lead to novel diagnostic tools and treatments for epilepsy. Applications may range from a warning to the patient to avert seizure-associated injuries, to automatic timely administration of an appropriate stimulus. Seizure prediction could become an integral part of the treatment of epilepsy through neuromodulation, especially in the new generation of closed-loop seizure control systems.
Epilepsy is a dynamical disorder with intermittent crises (seizures) that until recently were considered unpredictable. In this study, we investigated the predictability of epileptic seizures in chronically epileptic rats as a first step towards a subsequent timely intervention for seizure control. We look at the epileptic brain as a nonlinear complex system that undergoes spatio-temporal state transitions and the Lyapunov exponents as indices of its stability. We estimated the spatial synchronization or desynchronization of the maximum short-term Lyapunov exponents (STLmax, approximate measures of chaos) among multiple brain sites over days of electroencephalographic (EEG) recordings from 5 rats that had developed chronic epilepsy according to the lithium pilocarpine rodent model of epilepsy. We utilized this synchronization of EEG dynamics for the construction of a robust seizure prediction algorithm. The parameters of the algorithm were optimized using receiver operator curves (ROCs) on training EEG datasets from each rat for the algorithm to provide maximum sensitivity and specificity in the prediction of their seizures. The performance of the algorithm was then tested on long-term testing EEG datasets per rat. The thus optimized prediction algorithm on the testing datasets over all rats yielded a seizure prediction mean sensitivity of 85.9%, specificity of 0.180 false predictions per hour, and prediction time of 67.6 minutes prior to a seizure onset. This study provides evidence that prediction of seizures is feasible through analysis of the EEG within the framework of nonlinear dynamics, and thus paves the way for just-in-time pharmacological or physiological inter-ventions to abort seizures tens of minutes before their occurrence.
In the present paper a number of techniques were applied to determine the effects of epileptic seizure on spontaneous ongoing EEG. The idea is that seizure represents transitions of an epileptic brain from its normal (chaotic) state to an abnormal (more ordered) state. Some nonlinear measures including correlation dimension, maximum Lyapunov exponent and wavelet entropy and a graphical tool, named recurrence plot, as well as a novel technique that collects some statistics of the state space organization were used to characterize interictal, preictal and ictal states and derivate a phase transition. The novelty of this work includes of introducing new types of indicators base upon some nonlinear features besides of proposing a new feature of point distribution in phase space. Our results show that (1) these three states are separable in 3-D feature space of nonlinear measures with a gradual decrease of their quantity in seizure evolution, (2) strong rhythmicity, which manifests in recurrence plots and recurrence quantification analysis measures, appears in dynamic while having entered into seizure and (3) different volumes of state space are occupied during each phase of epileptic disorder. The significance of the work is that this information is a step into the detection of a preictal state and consequently is helpful in the prediction and control of epileptic seizures.
In 2008, a group of clinicians, scientists, engineers, and industry representatives met to discuss advances in the application of engineering technologies to the diagnosis and treatment of patients with epilepsy. The presentations also provided a guide for further technological development, specifically in the evaluation of patients for epilepsy surgery, seizure onset detection and seizure prediction, intracranial treatment systems, and extracranial treatment systems. This article summarizes the discussions and demonstrates that cross-disciplinary interactions can catalyze collaborations between physicians and engineers to address and solve many of the pressing unmet needs in epilepsy.
None of the current epileptic seizure prediction methods can widely be accepted, due to their poor consistency in performance. In this work, we have developed a novel approach to analyze intracranial EEG data. The energy of the frequency band of 4–12 Hz is obtained by wavelet transform. A dynamic model is introduced to describe the process and a hidden variable is included. The hidden variable can be considered as indicator of seizure activities. The method of particle filter associated with a neural network is used to calculate the hidden variable. Six patients' intracranial EEG data are used to test our algorithm including 39 hours of ictal EEG with 22 seizures and 70 hours of normal EEG recordings. The minimum least square error algorithm is applied to determine optimal parameters in the model adaptively. The results show that our algorithm can successfully predict 15 out of 16 seizures and the average prediction time is 38.5 minutes before seizure onset. The sensitivity is about 93.75% and the specificity (false prediction rate) is approximately 0.09 FP/h. A random predictor is used to calculate the sensitivity under significance level of 5%. Compared to the random predictor, our method achieved much better performance.
We have designed and implemented an automated, just-in-time stimulation, seizure control method using a seizure prediction method from nonlinear dynamics coupled with deep brain stimulation in the centromedial thalamic nuclei in epileptic rats. A comparison to periodic stimulation, with identical stimulation parameters, was also performed. The two schemes were compared in terms of their efficacy in control of seizures, as well as their effect on synchronization of brain dynamics. The automated just-in-time (JIT) stimulation showed reduction of seizure frequency and duration in 5 of the 6 rats, with significant reduction of seizure frequency (>50%) in 33% of the rats. This constituted a significant improvement over the efficacy of the periodic control scheme in the same animals. Actually, periodic stimulation showed an increase of seizure frequency in 50% of the rats, reduction of seizure frequency in 3 rats and significant reduction in 1 rat. Importantly, successful seizure control was highly correlated with desynchronization of brain dynamics. This study provides initial evidence for the use of closed-loop feedback control systems in epileptic seizures combining methods from seizure prediction and deep brain stimulation.
Current epileptic seizure "prediction" algorithms are generally based on the knowledge of seizure occurring time and analyze the electroencephalogram (EEG) recordings retrospectively. It is then obvious that, although these analyses provide evidence of brain activity changes prior to epileptic seizures, they cannot be applied to develop implantable devices for diagnostic and therapeutic purposes. In this paper, we describe an adaptive procedure to prospectively analyze continuous, long-term EEG recordings when only the occurring time of the first seizure is known. The algorithm is based on the convergence and divergence of short-term maximum Lyapunov exponents (STLmax) among critical electrode sites selected adaptively. A warning of an impending seizure is then issued. Global optimization techniques are applied for selecting the critical groups of electrode sites. The adaptive seizure prediction algorithm (ASPA) was tested in continuous 0.76 to 5.84 days intracranial EEG recordings from a group of five patients with refractory temporal lobe epilepsy. A fixed parameter setting applied to all cases predicted 82% of seizures with a false prediction rate of 0.16/h. Seizure warnings occurred an average of 71.7 min before ictal onset. Similar results were produced by dividing the available EEG recordings into half training and testing portions. Optimizing the parameters for individual patients improved sensitivity (84% overall) and reduced false prediction rate (0.12/h overall). These results indicate that ASPA can be applied to implantable devices for diagnostic and therapeutic purposes.
For almost 40 years, neuroscientists thought that epileptic seizures began abruptly, just a few seconds before clinical attacks. There is now mounting evidence that seizures develop minutes to hours before clinical onset. This change in thinking is based on quantitative studies of long digital intracranial electroencephalographic (EEG) recordings from patients being evaluated for epilepsy surgery. Evidence that seizures can be predicted is spread over diverse sources in medical, engineering, and patent publications. Techniques used to forecast seizures include frequency-based methods, statistical analysis of EEG signals, non-linear dynamics (chaos), and intelligent engineered systems. Advances in seizure prediction promise to give rise to implantable devices able to warn of impending seizures and to trigger therapy to prevent clinical epileptic attacks. Treatments such as electrical stimulation or focal drug infusion could be given on demand and might eliminate side-effects in some patients taking antiepileptic drugs long term. Whether closed-loop seizure-prediction and treatment devices will have the profound clinical effect of their cardiological predecessors will depend on our ability to perfect these techniques. Their clinical efficacy must be validated in large-scale, prospective, controlled trials.
To develop a prospective method for optimizing seizure prediction, given an array of implanted electrodes and a set of candidate quantitative features computed at each contact location.
The method employs a genetic-based selection process, and then tunes a probabilistic neural network classifier to predict seizures within a 10 min prediction horizon. Initial seizure and interictal data were used for training, and the remaining IEEG data were used for testing. The method continues to train and learn over time.
Validation of these results over two workshop patients demonstrated a sensitivity of 100%, and 1.1 false positives per hour for Patient E, using a 2.4s block predictor, and a failure of the method on Patient B.
This study demonstrates a prospective, exploratory implementation of a seizure prediction method designed to adapt to individual patients with a wide variety of pre-ictal patterns, implanted electrodes and seizure types. Its current performance is limited likely by the small number of input channels and quantitative features employed in this study, and segmentation of the data set into training and testing sets rather than using all continuous data available.
This technique theoretically has the potential to address the challenge presented by the heterogeneity of EEG patterns seen in medication-resistant epilepsy. A more comprehensive implementation utilizing all electrode sites, a broader feature library, and automated multi-feature fusion will be required to fully judge the method's potential for predicting seizures.
Most of the methods for prediction of epilepsy recently reported in the literature are based on the evaluation of chaotic behavior of intracranial electroencephalographic (EEG) recordings. These recordings require intensive surgical operations to implant the electrodes within the brain which are hazardous to the patient. Here, we have developed a novel approach to quantify the dynamical changes of the brain using the scalp EEG. The scalp signals are preprocessed by means of an effective block-based blind source separation (BSS) technique to separate the underlying sources within the brain. The algorithm significantly removes the effect of eye blinking artifacts. An overlap window procedure has been incorporated in order to mitigate the inherent permutation problem of BSS and maintain the continuity of the estimated sources. Chaotic behavior of the underlying sources has then been evaluated by measuring the largest Lyapunov exponent. For our experiments, we provided twenty sets of simultaneous intracranial and scalp EEG recordings from twenty patients. The above recordings have been compared. Similar results were obtained when the intracranial electrodes recorded the electrical activity of the epileptic focus. Our preliminary results show a great improvement when the epileptic focus is not captured by the intracranial electrodes.
Epileptic seizures are manifestations of epilepsy, a serious brain dynamical disorder second only to strokes. Of the world's ∼50 million people with epilepsy, fully 1/3 have seizures that are not controlled by anti-convulsant medication. The field of seizure prediction, in which engineering technologies are used to decode brain signals and search for precursors of impending epileptic seizures, holds great promise to elucidate the dynamical mechanisms underlying the disorder, as well as to enable implantable devices to intervene in time to treat epilepsy. There is currently an explosion of interest in this field in academic centers and medical industry with clinical trials underway to test potential prediction and intervention methodology and devices for Food and Drug Administration (FDA) approval. This invited paper presents an overview of the application of signal processing methodologies based upon the theory of nonlinear dynamics to the problem of seizure prediction. Broader application of these developments to a variety of systems requiring monitoring, forecasting and control is a natural outgrowth of this field.
ResearchGate has not been able to resolve any references for this publication.