Electroencephalogram Time Series Research Articles

An Efficient Epilepsy Prediction Model on European Dataset with Model Evaluation Considering Seizure Types.

This paper develops a computationally efficient model for automatic patient-specific seizure prediction using a two-layer LSTM from multichannel intracranial electroencephalogram time-series data. We decrease the number of parameters by employing a smaller input size and fewer electrodes, thereby making the model a viable option for wearable and implantable devices. We test the proposed prediction model on 26 patients from the European iEEG dataset, which is the largest epileptic seizure dataset. We also apply an automatic preprocessing technique based on a common average reference to remove artifacts from this dataset. The simulation results show that the model with its simple structure in conjunction with the mean post-processing procedure performed the best, with an average AUC of 0.885. This study is the first that utilizes the European database for epilepsy prediction application and the first that analyzes the effect of the seizure type on the system performance and demonstrates that the seizure type has a considerable impact.

Detecting cognitive traits and occupational proficiency using EEG and statistical inference

Machine learning (ML) is widely used in classification tasks aimed at detecting various cognitive states or neurological diseases using noninvasive electroencephalogram (EEG) time series. However, successfully detecting specific cognitive skills in a healthy population, independent of subject, remains challenging. This study compared the subject-independent classification performance of three different pipelines: supervised and Riemann projections with logistic regression and handcrafted power spectral features with light gradient boosting machine (LightGBM). 128-channel EEGs were recorded from 26 healthy volunteers while they solved arithmetic, logical, and verbal tasks. The participants were divided into two groups based on their higher education and occupation: specialists in mathematics and humanities. The balanced accuracy of the education type was significantly above chance for all pipelines: 0.84–0.89, 0.85–0.88, and 0.86–0.88 for each type of task, respectively. All three pipelines allowed us to distinguish mathematical proficiency based on learning experience with different trade-offs between performance and explainability. Our results suggest that ML approaches could also be effective for recognizing individual cognitive traits using EEG.

Network of compression networks to extract useful information from multivariate time series

Abstract Compression networks are the result of a recently proposed method to transform univariate time series to a complex network representation by using a compression algorithm. We show how a network of compression networks can be constructed to capture relationships among multivariate time series. This network is a weighted graph with edge weights corresponding to how well the compression codewords of one time series compress another time series. Subgraphs of this network obtained by thresholding of the relative compression edge weights are shown to possess properties which can track dynamical change. Furthermore, community structures—groups of vertices more densely connected together—within these networks can identify partially synchronized states in the dynamics of networked oscillators, as well as perform genre classification of musical compositions. An additional example incorporates temporal windowing of the data and demonstrates the potential of the method to identify tipping point behaviour through the analysis of multivariate electroencephalogram time series of patients undergoing seizure.

An intelligent optimized deep learning model to achieve early prediction of epileptic seizures

Seizure prediction from electroencephalogram (EEG) time series data and a sequential deep learning (DL) predictor substantially boosts epileptic patients’ quality of life. However, a significant challenge is a variation in seizure characteristics with time and individuals along with a need for more data. Also, considerable dissimilarity is noticed in the duration between various seizure stages. Thus, a patient-generic approach is required to mitigate the problem. As a result, multiple feature augmentation procedures are used to create a hybrid feature space to capture the non-linearity of epileptic seizures. This elaborate feature space helps the predictor learn better to enhance the seizure prediction. Additionally, the predictor is optimized using a novel hybrid Forensic-based-Search-and-Rescue Optimization (FB-SARO) to improve the seizure prediction. In addition, an optimal seizure prediction horizon (SPH) is also determined through the classifier’s learning. The SPH helps attain early prediction while preserving accuracy and achieving a minimum False Prediction Rate (FPR). It also helps raise the alarm to provide the patients with ample preparation time for medical assistance. The proposed approach is testified through publicly available datasets and compared with existing state-of-the-art techniques.

Effects of sampling rate on multiscale entropy of electroencephalogram time series

A physiological system encompasses numerous components that function at various time scales. To characterize the scale-dependent feature, the multiscale entropy (MSE) analysis has been proposed to describe the complex processes on multiple time scales. However, MSE analysis uses the relative scale factors to reveal the time-related dynamics, which may cause in-comparability of results from diverse studies with inconsistent sampling rates. In this study, in addition to the conventional MSE with relative scale factors, we also expressed MSE with absolute time scales (MaSE). We compared the effects of sampling rates on MSE and MaSE of simulated and real EEG time series. The results show that the previously found phenomenon (down-sampling can increase sample entropy) is just the projection of the compressing effect of down-sampling on MSE. And we have also shown the compressing effect of down-sampling on MSE does not change MaSE’s profile, despite some minor right-sliding. In addition, by analyzing a public EEG dataset of emotional states, we have demonstrated improved classification rate after choosing appropriate sampling rate. We have finally proposed a working strategy to choose an appropriate sampling rate, and suggested using MaSE to avoid confusion caused by sampling rate inconsistency. This novel study may apply to a broad range of studies that would traditionally utilize sample entropy and MSE to analyze the complexity of an underlying dynamic process.

Introducing block-Toeplitz covariance matrices to remaster linear discriminant analysis for event-related potential brain–computer interfaces

Objective. Covariance matrices of noisy multichannel electroencephalogram (EEG) time series data provide essential information for the decoding of brain signals using machine learning methods. However, small datasets and high dimensionality make it hard to estimate these matrices. In brain–computer interfaces (BCI) based on event-related potentials (ERP) and a linear discriminant analysis (LDA) classifier, the state of the art covariance estimation uses shrinkage regularization. As this is a general covariance regularization approach, we aim at improving LDA further by better exploiting the domain-specific characteristics of the EEG to regularize the covariance estimates. Approach. We propose to enforce a block-Toeplitz structure for the covariance matrix of the LDA, which implements an assumption of signal stationarity in short time windows for each channel. Main results. An offline re-analysis of data collected from 213 subjects under 13 different event-related potential BCI protocols showed a significantly increased binary classification performance of this ‘ToeplitzLDA’ compared to shrinkage regularized LDA (up to 6 AUC points, p < 0.001) and Riemannian classification approaches (up to 2 AUC points, p < 0.001). In an unsupervised visual speller application, this improvement would translate to a relative reduction of spelling errors by 81% on average for 25 subjects. Additionally, aside from lower memory and reduced time complexity for LDA training, ToeplitzLDA proves to be robust against drastic increases of the number of temporal features. Significance. The proposed covariance estimation allows BCI researchers to improve classification rates and reduce calibration times of BCI protocols using event-related potentials and thus support the usability of corresponding applications. Its lower computational and memory needs could make it a valuable algorithm especially for mobile BCIs.

A Super Fast Algorithm for Estimating Sample Entropy.

Sample entropy, an approximation of the Kolmogorov entropy, was proposed to characterize complexity of a time series, which is essentially defined as , where B denotes the number of matched template pairs with length m and A denotes the number of matched template pairs with , for a predetermined positive integer m. It has been widely used to analyze physiological signals. As computing sample entropy is time consuming, the box-assisted, bucket-assisted, x-sort, assisted sliding box, and kd-tree-based algorithms were proposed to accelerate its computation. These algorithms require or computational complexity, where N is the length of the time series analyzed. When N is big, the computational costs of these algorithms are large. We propose a super fast algorithm to estimate sample entropy based on Monte Carlo, with computational costs independent of N (the length of the time series) and the estimation converging to the exact sample entropy as the number of repeating experiments becomes large. The convergence rate of the algorithm is also established. Numerical experiments are performed for electrocardiogram time series, electroencephalogram time series, cardiac inter-beat time series, mechanical vibration signals (MVS), meteorological data (MD), and noise. Numerical results show that the proposed algorithm can gain 100–1000 times speedup compared to the kd-tree and assisted sliding box algorithms while providing satisfactory approximate accuracy.

Classification of Epilepsy Types from Electroencephalogram Time Series Using Continuous Wavelet Transform Scalogram–Based Convolutional Neural Network

Abstract During the supervisory activities of the brain, the electrical activities of nerve cell clusters produce oscillations. These complex biopotential oscillations are called electroencephalogram (EEG) signals. Certain diseases, such as epilepsy, can be detected by measuring these signals. Epilepsy is a disease that manifests itself as seizures. These seizures manifest themselves in different characteristics. These different characteristics divide epilepsy seizure types into two main groups: generalized and partial epilepsy. This study aimed to classify different types of epilepsy from EEG signals. For this purpose, a scalogram-based, deep learning approach has been developed. The utilized classification process had the following main steps: the scalogram images were obtained by using the continuous wavelet transform (CWT) method. So, a one-dimension EEG time series was converted to a two-dimensional time-frequency data set in order to extract more features. Then, the increased dimension data set (CWT scalogram images) was applied to the convolutional neural network (CNN) as input patterns for classifying the images. The EEG signals were taken from Dicle University, Neurology Clinic of Medical School. This data consisted of four classes: healthy brain waves, generalized preseizure, generalized seizure, and partial epilepsy brain waves. With the proposed method, the average accuracy performance of three of the EEG records’ classes (healthy, generalized preseizure, and generalized seizure), and that of all four classes of EEG records were 90.16 % (± 0.20) and 84.66 % (± 0.48). According to these results, regarding the specific accuracy ratings of the recordings, the healthy EEG records scored 91.29 %, generalized epileptic seizure records were at 96.50 %, partial seizure EEG records scored 89.63 %, and the preseizure EEG records had a 90.44 % rating. The results of the proposed method were compared to the results of both similar studies and conventional methods. As a result, the performance of the proposed method was found to be acceptable.

Closeness-centrality-correlation for detecting interdependency between coupled systems

Generalized synchronization is a common interdependency between coupled systems which exists in many branches of life, social and physical science. In this paper, a novel method, called closeness-centrality-correlation is proposed for the detection of this interdependency. The proposed method is based on a global network measure (i.e., closeness centrality) of recurrence networks resulting from time series. We illustrate the feasibility of the proposed method using a paradigmatic coupled model and compare its performance to other commonly used interdependency methods. The numerical results show that the proposed method is quite satisfactory for detecting interdependency and outperforms the existing joint probability of recurrence method especially for the case that the dynamics of the two coupled subsystems are significantly different. Moreover, through analyzing the time series contaminated by white noise, we demonstrate that our method is robust against white noise. Finally, an application to recorded electroencephalogram data shows that the proposed measure is more reliable to detect the transitions of the interdependencies among the noisy electroencephalogram time series and thus provides longer pre-warning time for the onset of epilepsy.

A Complex-Valued Oscillatory Neural Network for Storage and Retrieval of Multidimensional Aperiodic Signals.

Recurrent neural networks with associative memory properties are typically based on fixed-point dynamics, which is fundamentally distinct from the oscillatory dynamics of the brain. There have been proposals for oscillatory associative memories, but here too, in the majority of cases, only binary patterns are stored as oscillatory states in the network. Oscillatory neural network models typically operate at a single/common frequency. At multiple frequencies, even a pair of oscillators with real coupling exhibits rich dynamics of Arnold tongues, not easily harnessed to achieve reliable memory storage and retrieval. Since real brain dynamics comprises of a wide range of spectral components, there is a need for oscillatory neural network models that operate at multiple frequencies. We propose an oscillatory neural network that can model multiple time series simultaneously by performing a Fourier-like decomposition of the signals. We show that these enhanced properties of a network of Hopf oscillators become possible by operating in the complex-variable domain. In this model, the single neural oscillator is modeled as a Hopf oscillator, with adaptive frequency and dynamics described over the complex domain. We propose a novel form of coupling, dubbed “power coupling,” between complex Hopf oscillators. With power coupling, expressed naturally only in the complex-variable domain, it is possible to achieve stable (normalized) phase relationships in a network of multifrequency oscillators. Network connections are trained either by Hebb-like learning or by delta rule, adapted to the complex domain. The network is capable of modeling N-channel electroencephalogram time series with high accuracy and shows the potential as an effective model of large-scale brain dynamics.

Unsupervised Learning in Reservoir Computing for EEG-Based Emotion Recognition

In real-world applications such as emotion recognition from recorded brain activity, data are captured from electrodes over time. These signals constitute a multidimensional time series. In this article, Echo State Network (ESN), a recurrent neural network with great success in time series prediction and classification, is optimized with different neural plasticity rules for classification of emotions based on electroencephalogram (EEG) time series. The developed network could automatically extract valid features from EEG signals. We use the filtered signals as the network input and do not take any feature extraction methods. Evaluated on two well-known benchmarks, the DEAP dataset, and the SEED dataset, the performance of the ESN with intrinsic plasticity greatly outperforms the feature-based methods and shows certain advantages compared with other existing methods. Thus, the proposed network can form a more complete and efficient representation, whilst retaining the advantages such as faster learning speed and more reliable performance.

Early prediction of epileptic seizures using a long-term recurrent convolutional network

BackgroundA seizure prediction system can detect seizures prior to their occurrence and allow clinicians to provide timely treatment for patients with epilepsy. Research on seizure prediction has progressed from signal processing analyses to machine learning. However, most prediction methods are hand-engineered and have high computational complexity, increasing the difficulty of obtaining real-time predictions. Some forecasting and early warning methods have achieved good results in the short term but have low applicability in practical situations over the long term. New methodsFirst, electroencephalogram (EEG) time series were converted into two-dimensional images for multichannel fusion. A feasible method, a long-term recurrent convolutional network (LRCN), was proposed to create a spatiotemporal deep learning model for predicting epileptic seizures. The convolutional network block was used to automatically extract deep features from the data. The long short-term memory (LSTM) block was incorporated into learning a time sequence for identifying the preictal segments. New network settings and a postprocessing strategy were proposed in the seizure prediction model. ResultsThe deep seizure prediction model achieved an accuracy of 93.40%, prediction sensitivity of 91.88% and specificity of 86.13% in segment-based evaluations. For the event-based evaluations, 164 seizures were predicted. The proposed method provides high sensitivity and a low false prediction rate (FPR) of 0.04 F P/h. Comparison with existing methodsWe employed different methods, including the LRCN, deep learning and traditional machine learning methods, and compared them using the same data in this paper. Overall, the LRCN offers approximately 5–9% increased sensitivity and specificity. ConclusionThis study describes the LRCN network for analyzing EEG data to predict epileptic seizures, thereby enabling the implementation of an early warning system that detects epileptic seizures before they occur in clinical applications.

Classification of Hand Movements From EEG Using a Deep Attention-Based LSTM Network

Classifying limb movements using brain activity is an important task in Brain-computer Interfaces (BCI) that has been successfully used in multiple application domains, ranging from human-computer interaction to medical and biomedical applications. This paper proposes a novel solution for classification of left/right hand movement by exploiting a Long Short-Term Memory (LSTM) network with attention mechanism to learn the electroencephalogram (EEG) time-series information. To this end, a wide range of time and frequency domain features are extracted from the EEG signals and used to train an LSTM network to perform the classification task. We conduct extensive experiments with the EEG Movement dataset and show that our proposed solution our method achieves improvements over several benchmarks and state-of-the-art methods in both intra-subject and cross-subject validation schemes. Moreover, we utilize the proposed framework to analyze the information as received by the sensors and monitor the activated regions of the brain by tracking EEG topography throughout the experiments.

Quantifying interdependence using the missing joint ordinal patterns.

In this paper, we develop the concept of forbidden/missing ordinal patterns into the forbidden/missing joint ordinal patterns and propose the ratio of the number of missing joint ordinal patterns (RMJPs) as a sign of interdependence. RMJP in a surrogate analysis can be used to differentiate the forbidden joint ordinal patterns from the missing joint ordinal patterns due to small sample effects. We first apply RMJP to the simulated time series: a two-component autoregressive fractionally integrated moving average process, the Hénon map, and the Rössler system using active control and discuss the effect of the length of the time series, embedding dimension, and noise contamination. RMJP has been proven to be capable of measuring the interdependence in the numerical simulation. Then, RMJP is further used on the electroencephalogram (EEG) time series for empirical analysis to explore the interdependence of brain waves. With results by RMJP obtained from a widely used open dataset of the sleep EEG time series from healthy subjects, we find that RMJP can be used to quantify the brain wave interdependence under different sleep/wake stages, reveal the overall sleep architecture, and indicate a higher level of interdependence as sleep gets deeper. The findings are consistent with existing knowledge in sleep medicine. The proposed RMJP method has shown its validity and applicability and may assist automatic sleep quantification or bring insight into the understanding of the brain activity during sleep. Furthermore, RMJP can be used on sleep EEG under various pathological conditions and in large-scale sleep studies, helping to investigate the mechanisms of the sleep process and neuron synchronization.

Prediction of Electroencephalogram Time Series With Electro-Search Optimization Algorithm Trained Adaptive Neuro-Fuzzy Inference System

Nowadays, artificial intelligence is widely used in many biomedical-oriented problems. Because of obtained effective and efficient results, the use of intelligent solution mechanisms by artificial intelligence techniques is mainly focused on healthcare applications. Moving from the explanations, the objective of this paper is to provide an adaptive neuro-fuzzy inference system (ANFIS) trained by a recent optimization algorithm: electro-search optimization (ESO) for predicting the electroencephalogram (EEG) time series. In detail, the research was directed to the EEG time series showing chaotic characteristics so that an effective hybrid system can be designed for having an idea about future states of human brain activity in the case of possible diseases. The developed ANFIS-ESO system was evaluated with five different EEG time series and the obtained findings were reported accordingly. In addition, the ANFIS-ESO system was compared with alternative techniques-systems in order to see the performances according to different systems. In the end, it is possible to mention that the ANFIS-ESO system provides well-enough results in terms of predicting EEG time series. As a result of encouraging results, ANFIS-ESO is currently used actively for real cases.

Time-time analysis of electroencephalogram signals for epileptic seizure detection

The detection and classification of epileptic seizures using the electroencphalogram (EEG) signals has been actively worked upon by the researchers from past few decades. This paper attempts a novel application of time-time transform for analysis of electroencephalogram time-series for epileptic seizure detection by transforming it into secondary time-limited local constituent time-series. This technique (TT-transform) of time-time representation of the EEG time series is derived from S-transform, i.e., Stockwell transform (an extension of the wavelets), a method that represents a non-stationary time series as a set of complex time-localised spectra. With the help of TT-transform, a more informative representation of the time features of EEG signals has been obtained, around a particular point on the time axis which has been seen to prove very effective in seizure detection. As the TT-transform is completely invertible, it indicates frequency filtering and signal to noise improvements in the time domain. Features obtained upon application of TT-transform on EEG time-series are classified using quadratic discriminant analysis and the correct classification rate obtained is 100%.

Constructing brain connectivity group graphs from EEG time series

ABSTRACTGraphical analysis of complex brain networks is a fundamental area of modern neuroscience. Functional connectivity is important since many neurological and psychiatric disorders, including schizophrenia, are described as ‘dys-connectivity’ syndromes. Using electroencephalogram time series collected on each of a group of 15 individuals with a common medical diagnosis of positive syndrome schizophrenia we seek to build a single, representative, brain functional connectivity group graph. Disparity/distance measures between spectral matrices are identified and used to define the normalized graph Laplacian enabling clustering of the spectral matrices for detecting ‘outlying’ individuals. Two such individuals are identified. For each remaining individual, we derive a test for each edge in the connectivity graph based on average estimated partial coherence over frequencies, and associated p-values are found. For each edge these are used in a multiple hypothesis test across individuals and the proportion rejecting the hypothesis of no edge is used to construct a connectivity group graph. This study provides a framework for integrating results on multiple individuals into a single overall connectivity structure.

Multichannel EEG based inter-ictal seizures detection using Teager energy with backpropagation neural network classifier.

A long-term multichannel electroencephalogram recording plays a crucial role in recognizing the epileptic seizure activities from the brain lobes. This research study investigates the automated detection of epileptic seizures from multichannel electroencephalogram recordings using Teager energy feature. A supervised back-propagation neural network model was implemented to classify the inter-ictal seizures. The study was conducted on multichannel electroencephalogram data that was obtained from Institute of Neuroscience, Ramaiah Memorial Hospital, Bengaluru, India, after ethical clearance from the from the Institutional Ethics Board. Initially, notch filter was applied to remove the 50Hz power line noise from raw electroencephalogram followed by independent component analysis to remove eye blinks and muscular activities. A time domain feature called Teager energy was estimated which detects the rapid changes in the given electroencephalogram time series. A 1s windowing was introduced to ensure stationarity for estimation of Teager energy. The descriptive and box plot analysis ensures the suitability of the Teager energy for the seizure detection. The performance of the multilayer perceptron neural network classifier was evaluated using sensitivity, specificity, and false detection rate. Simulation results showed the highest sensitivity, specificity and false detection rate of 96.66%, 99.15%, and 0.30 per hour respectively. It can be concluded that procedure can be applied for real-time seizure detection.

An Ant-Lion Optimizer-Trained Artificial Neural Network System for Chaotic Electroencephalogram (EEG) Prediction

The prediction of future events based on available time series measurements is a relevant research area specifically for healthcare, such as prognostics and assessments of intervention applications. A measure of brain dynamics, electroencephalogram time series, are routinely analyzed to obtain information about current, as well as future, mental states, and to detect and diagnose diseases or environmental factors. Due to their chaotic nature, electroencephalogram time series require specialized techniques for effective prediction. The objective of this study was to introduce a hybrid system developed by artificial intelligence techniques to deal with electroencephalogram time series. Both artificial neural networks and the ant-lion optimizer, which is a recent intelligent optimization technique, were employed to comprehend the related system and perform some prediction applications over electroencephalogram time series. According to the obtained findings, the system can successfully predict the future states of target time series and it even outperforms some other hybrid artificial neural network-based systems and alternative time series prediction approaches from the literature.

Bodily-Contact Communication Medium Induces Relaxed Mode of Brain Activity While Increasing Its Dynamical Complexity: A Pilot Study.

We present the results of the analysis of the effect of a bodily-contact communication medium on the brain activity of the individuals during verbal communication. Our results suggest that the communicated content that is mediated through such a device induces a significant effect on electroencephalogram (EEG) time series of human subjects. Precisely, we find a significant reduction of overall power of the EEG signals of the individuals. This observation that is supported by the analysis of the permutation entropy (PE) of the EEG time series of brain activity of the participants suggests the positive effect of such a medium on the stress relief and the induced sense of relaxation. Additionally, multiscale entropy (MSE) analysis of our data implies that such a medium increases the level of complexity that is exhibited by EEG time series of our participants, thereby suggesting their sustained sense of involvement in their course of communication. These findings that are in accord with the results reported by cognitive neuroscience research suggests that the use of such a medium can be beneficial as a complementary step in treatment of developmental disorders, attentiveness of schoolchildren and early child development, as well as scenarios where intimate physical interaction over distance is desirable (e.g., distance-parenting).