Artificial intelligence in epilepsy - applications and pathways to the clinic

Artificial intelligence (AI) is rapidly transforming health care, and its applications in epilepsy have increased exponentially over the past decade. Integration of AI into epilepsy management promises to revolutionize the diagnosis and treatment of this complex disorder. However, translation of AI into neurology clinical practice has not yet been successful, emphasizing the need to consider progress to date and assess challenges and limitations of AI. In this Review, we provide an overview of AI applications that have been developed in epilepsy using a variety of data modalities: neuroimaging, electroencephalography, electronic health records, medical devices and multimodal data integration. For each, we consider potential applications, including seizure detection and prediction, seizure lateralization, localization of the seizure-onset zone and assessment for surgical or neurostimulation interventions, and review the performance of AI tools developed to date. We also discuss methodological considerations and challenges that must be addressed to successfully integrate AI into clinical practice. Our goal is to provide an overview of the current state of the field and provide guidance for leveraging AI in future to improve management of epilepsy.

Effective Early Detection of Epileptic Seizures through EEG Signals Using Classification Algorithms Based on t-Distributed Stochastic Neighbor Embedding and K-Means

Epilepsy is a neurological disorder in the activity of brain cells that leads to seizures. An electroencephalogram (EEG) can detect seizures as it contains physiological information of the neural activity of the brain. However, visual examination of EEG by experts is time consuming, and their diagnoses may even contradict each other. Thus, an automated computer-aided diagnosis for EEG diagnostics is necessary. Therefore, this paper proposes an effective approach for the early detection of epilepsy. The proposed approach involves the extraction of important features and classification. First, signal components are decomposed to extract the features via the discrete wavelet transform (DWT) method. Principal component analysis (PCA) and the t-distributed stochastic neighbor embedding (t-SNE) algorithm were applied to reduce the dimensions and focus on the most important features. Subsequently, K-means clustering + PCA and K-means clustering + t-SNE were used to divide the dataset into subgroups to reduce the dimensions and focus on the most important representative features of epilepsy. The features extracted from these steps were fed to extreme gradient boosting, K-nearest neighbors (K-NN), decision tree (DT), random forest (RF) and multilayer perceptron (MLP) classifiers. The experimental results demonstrated that the proposed approach provides superior results to those of existing studies. During the testing phase, the RF classifier with DWT and PCA achieved an accuracy of 97.96%, precision of 99.1%, recall of 94.41% and F1 score of 97.41%. Moreover, the RF classifier with DWT and t-SNE attained an accuracy of 98.09%, precision of 99.1%, recall of 93.9% and F1 score of 96.21%. In comparison, the MLP classifier with PCA + K-means reached an accuracy of 98.98%, precision of 99.16%, recall of 95.69% and F1 score of 97.4%.

Classification of focal and non-focal EEG signals using optimal geometrical features derived from a second-order difference plot of FBSE-EWT rhythms

BACKGROUND/INTRODUCTION

Manual detection and localization of the brain's epileptogenic areas using electroencephalogram (EEG) signals is time-intensive and error-prone. An automated detection system is, thus, highly desirable for support in clinical diagnosis. A set of relevant and significant non-linear features plays a major role in developing a reliable, automated focal detection system.

METHODS

A new feature extraction method is designed to classify focal EEG signals using eleven non-linear geometrical attributes derived from the Fourier-Bessel series expansion-based empirical wavelet transform (FBSE-EWT) segmented rhythm's second-order difference plot (SODP). A total of 132 features (2 channels × 6 rhythms × 11 geometrical attributes) were computed. However, some of the obtained features might be non-significant and redundant features. Hence, to acquire an optimal set of relevant non-linear features, a new hybridization of 'Kruskal-Wallis statistical test (KWS)' with 'VlseKriterijuska Optimizacija I Komoromisno Resenje' termed as the KWS-VIKOR approach was adopted. The KWS-VIKOR has a two-fold operational feature. First, the significant features are selected using the KWS test with a p-value lesser than 0.05. Next, the multi-attribute decision-making (MADM) based VIKOR method ranks the selected features. Several classification methods further validate the efficacy of the features of the selected top n%.

RESULTS

The proposed framework has been evaluated using the Bern-Barcelona dataset. The highest classification accuracy of 98.7% was achieved using the top 35% ranked features in classifying the focal and non-focal EEG signals with the least-squares support vector machine (LS-SVM) classifier.

CONCLUSIONS

The achieved results exceeded those reported through other methods. Hence, the proposed framework will more effectively assist the clinician in localizing the epileptogenic areas.

Epileptic seizure focus detection from interictal electroencephalogram: a survey

Electroencephalogram (EEG) is one of most effective clinical diagnosis modalities for the localization of epileptic focus. Most current AI solutions use this modality to analyze the EEG signals in an automated manner to identify the epileptic seizure focus. To develop AI system for identifying the epileptic focus, there are many recently-published AI solutions based on biomarkers or statistic features that utilize interictal EEGs. In this review, we survey these solutions and find that they can be divided into three main categories: (i) those that use of biomarkers in EEG signals, including high-frequency oscillation, phase-amplitude coupling, and interictal epileptiform discharges, (ii) others that utilize feature-extraction methods, and (iii) solutions based upon neural networks (an end-to-end approach). We provide a detailed description of seizure focus with clinical diagnosis methods, a summary of the public datasets that seek to reduce the research gap in epilepsy, recent novel performance evaluation criteria used to evaluate the AI systems, and guidelines on when and how to use them. This review also suggests a number of future research challenges that must be overcome in order to design more efficient computer-aided solutions to epilepsy focus detection.

Source-sink connectivity: a novel interictal EEG marker for seizure localization

Over 15 million epilepsy patients worldwide have drug-resistant epilepsy. Successful surgery is a standard of care treatment but can only be achieved through complete resection or disconnection of the epileptogenic zone, the brain region(s) where seizures originate. Surgical success rates vary between 20% and 80%, because no clinically validated biological markers of the epileptogenic zone exist. Localizing the epileptogenic zone is a costly and time-consuming process, which often requires days to weeks of intracranial EEG (iEEG) monitoring. Clinicians visually inspect iEEG data to identify abnormal activity on individual channels occurring immediately before seizures or spikes that occur interictally (i.e. between seizures). In the end, the clinical standard mainly relies on a small proportion of the iEEG data captured to assist in epileptogenic zone localization (minutes of seizure data versus days of recordings), missing opportunities to leverage these largely ignored interictal data to better diagnose and treat patients. IEEG offers a unique opportunity to observe epileptic cortical network dynamics but waiting for seizures increases patient risks associated with invasive monitoring. In this study, we aimed to leverage interictal iEEG data by developing a new network-based interictal iEEG marker of the epileptogenic zone. We hypothesized that when a patient is not clinically seizing, it is because the epileptogenic zone is inhibited by other regions. We developed an algorithm that identifies two groups of nodes from the interictal iEEG network: those that are continuously inhibiting a set of neighbouring nodes ('sources') and the inhibited nodes themselves ('sinks'). Specifically, patient-specific dynamical network models were estimated from minutes of iEEG and their connectivity properties revealed top sources and sinks in the network, with each node being quantified by source-sink metrics. We validated the algorithm in a retrospective analysis of 65 patients. The source-sink metrics identified epileptogenic regions with 73% accuracy and clinicians agreed with the algorithm in 93% of seizure-free patients. The algorithm was further validated by using the metrics of the annotated epileptogenic zone to predict surgical outcomes. The source-sink metrics predicted outcomes with an accuracy of 79% compared to an accuracy of 43% for clinicians' predictions (surgical success rate of this dataset). In failed outcomes, we identified brain regions with high metrics that were untreated. When compared with high frequency oscillations, the most commonly proposed interictal iEEG feature for epileptogenic zone localization, source-sink metrics outperformed in predictive power (by a factor of 1.2), suggesting they may be an interictal iEEG fingerprint of the epileptogenic zone.

A Classification Model of EEG Signals Based on RNN-LSTM for Diagnosing Focal and Generalized Epilepsy

Epilepsy is a chronic neurological disorder caused by abnormal neuronal activity that is diagnosed visually by analyzing electroencephalography (EEG) signals.

BACKGROUND

Surgical operations are the only option for epilepsy treatment when patients are refractory to treatment, which highlights the role of classifying focal and generalized epilepsy syndrome. Therefore, developing a model to be used for diagnosing focal and generalized epilepsy automatically is important.

METHODS

A classification model based on longitudinal bipolar montage (LB), discrete wavelet transform (DWT), feature extraction techniques, and statistical analysis in feature selection for RNN combined with long short-term memory (LSTM) is proposed in this work for identifying epilepsy. Initially, normal and epileptic LB channels were decomposed into three levels, and 15 various features were extracted. The selected features were extracted from each segment of the signals and fed into LSTM for the classification approach.

RESULTS

The proposed algorithm achieved a 96.1% accuracy, a 96.8% sensitivity, and a 97.4% specificity in distinguishing normal subjects from subjects with epilepsy. This optimal model was used to analyze the channels of subjects with focal and generalized epilepsy for diagnosing purposes, relying on statistical parameters.

CONCLUSIONS

The proposed approach is promising, as it can be used to detect epilepsy with satisfactory classification performance and diagnose focal and generalized epilepsy.

Automatic detection of abnormal EEG signals using multiscale features with ensemble learning

Electroencephalogram (EEG) is an economical and convenient auxiliary test to aid in the diagnosis and analysis of brain-related neurological diseases. In recent years, machine learning has shown great potential in clinical EEG abnormality detection. However, existing methods usually fail to consider the issue of feature redundancy when extracting the relevant EEG features. In addition, the importance of utilizing the patient age information in EEG detection is ignored. In this paper, a new framework is proposed for distinguishing an unknown EEG recording as either normal or abnormal by identifying different types of EEG-derived significant features. In the proposed framework, different hierarchical salient features are extracted using a time-wise multi-scale aggregation strategy, based on a selected group of statistical characteristics calculated from the optimum discrete wavelet transform coefficients. We also fuse the age information with multi-scale features for further improving discrimination. The integrated features are classified using three ensemble learning classifiers, CatBoost, LightGBM, and random forest. Experimental results show that our method with CatBoost classifier can yield superior performance vis-a-vis competing techniques, which indicates the great promise of our methodology in EEG pathology detection.

An Anomaly Detection Method of Time Series Data for Cyber-Physical Integrated Energy System Based on Time-Frequency Feature Prediction

An integrated energy system (IES) is vulnerable to network attacks due to the coupling features of multi-energy systems, as well as the deep integration between a physical system and an information system. The anomaly detection of the time-series data in an IES is a key problem to defend against network attacks and ensure the cyber-physical security of IES. Aiming at false data injection attacks (FDIAs) on IES, this paper proposes an anomaly detection method for time-series data in a cyber-physical integrated energy system based on time-frequency feature prediction. The time-frequency features of the time-series data are extracted based on three time-frequency transform methods (DWT, EMD, and EWT). Then the extracted time-frequency features are input to the autoencoder (AE) to capture the hidden features and nonlinear structure of the original data in the frequency domain. The time-domain data within the detected time period are predicted by applying regression prediction on the top-layer features of AE. Considering the uncertainty of regression prediction, kernel density estimation (KDE) is used to estimate the probability density function of prediction error and the interval of the predicted data is estimated accordingly. The estimated lower boundary value of the predicted data is selected as the attack judgment threshold for anomaly detection. The results of the case study verify the advantages of the proposed method in reducing the false positive rate and improving the anomaly detection accuracy.

Decoding Intracranial EEG With Machine Learning: A Systematic Review

Advances in intracranial electroencephalography (iEEG) and neurophysiology have enabled the study of previously inaccessible brain regions with high fidelity temporal and spatial resolution. Studies of iEEG have revealed a rich neural code subserving healthy brain function and which fails in disease states. Machine learning (ML), a form of artificial intelligence, is a modern tool that may be able to better decode complex neural signals and enhance interpretation of these data. To date, a number of publications have applied ML to iEEG, but clinician awareness of these techniques and their relevance to neurosurgery, has been limited. The present work presents a review of existing applications of ML techniques in iEEG data, discusses the relative merits and limitations of the various approaches, and examines potential avenues for clinical translation in neurosurgery. One-hundred-seven articles examining artificial intelligence applications to iEEG were identified from 3 databases. Clinical applications of ML from these articles were categorized into 4 domains: i) seizure analysis, ii) motor tasks, iii) cognitive assessment, and iv) sleep staging. The review revealed that supervised algorithms were most commonly used across studies and often leveraged publicly available timeseries datasets. We conclude with recommendations for future work and potential clinical applications.

Extracting Epileptic Features in EEGs Using a Dual-Tree Complex Wavelet Transform Coupled with a Classification Algorithm

The detection of epileptic seizures from electroencephalogram (EEG) signals is traditionally performed by clinical experts through visual inspection. It is a long process, is error prone, and requires a highly trained expert. In this research, a new method is presented for seizure classification for EEG signals using a dual-tree complex wavelet transform (DT-CWT) and fast Fourier transform (FFT) coupled with a least square support vector machine (LS-SVM) classifier. In this method, each EEG signal is divided into four segments. Each segment is further split into smaller sub-segments. The DT-CWT is applied to decompose each sub-segment into detailed and approximation coefficients (real and imaginary parts). The obtained coefficients by the DT-CWT at each decomposition level are passed through an FFT to identify the relevant frequency bands. Finally, a set of effective features are extracted from the sub-segments, and are then forwarded to the LS-SVM classifier to classify epileptic EEGs. In this paper, two epileptic EEG databases from Bonn and Bern Universities are used to evaluate the extracted features using the proposed method. The experimental results demonstrate that the method obtained an average accuracy of 97.7% and 96.8% for the Bonn and Bern databases, respectively. The results prove that the proposed DT-CWT and FFT based features extraction is an effective way to extract discriminative information from brain signals. The obtained results are also compared to those by k-means and Naïve Bayes classifiers as well as with the results from the previous methods reported for classifying epileptic seizures and identifying the focal and non-focal EEG signals. The obtained results show that the proposed method outperforms the others and it is effective in detecting epileptic seziures in EEG signals. The technique can be adopted to aid neurologists to better diagnose neurological disorders and for an early seizure warning system.

Exploiting Feature Selection and Neural Network Techniques for Identification of Focal and Nonfocal EEG Signals in TQWT Domain

For drug resistance patients, removal of a portion of the brain as a cause of epileptic seizures is a surgical remedy. However, before surgery, the detailed analysis of the epilepsy localization area is an essential and logical step. The Electroencephalogram (EEG) signals from these areas are distinct and are referred to as focal, while the EEG signals from other normal areas are known as nonfocal. The visual inspection of multiple channels for detecting the focal EEG signal is time-consuming and prone to human error. To address this challenge, we propose a novel method based on differential operator and Tunable Q-factor wavelet transform (TQWT) to distinguish the focal and nonfocal signals. For this purpose, first, the EEG signal was differenced and then decomposed by TQWT. Second, several entropy-based features were derived from the TQWT subbands. Third, the efficacy of the six binary feature selection algorithms, binary bat algorithm (BBA), binary differential evolution (BDE) algorithm, firefly algorithm (FA), genetic algorithm (GA), grey wolf optimization (GWO), and particle swarm optimization (PSO), was evaluated. In the end, the selected features were fed to several machine learning and neural network classifiers. We observed that the PSO with neural networks provides an effective solution for the application of focal EEG signal detection. The proposed framework resulted in an average classification accuracy of 97.68%, a sensitivity of 97.26%, and a specificity of 98.11% in a tenfold cross-validation strategy, which is higher than the state of the art used in the public Bern-Barcelona EEG database.

A Wavelet-Based Learning Model Enhances Molecular Prognosis in Pancreatic Adenocarcinoma

Genome-wide omics technology boosts deep interrogation into the clinical prognosis and inherent mechanism of pancreatic oncology. Classic LASSO methods coequally treat all candidates, ignoring individual characteristics, thus frequently deteriorating performance with comparatively more predictors. Here, we propose a wavelet-based deep learning method in variable selection and prognosis formulation for PAAD with small samples and multisource information. With the genomic, epigenomic, and clinical cohort information from The Cancer Genome Atlas, the constructed five-molecule model is validated via Kaplan-Meier survival estimate, rendering significant prognosis capability on high- and low-risk subcohorts (p value < 0.0001), together with three predictors manifesting the individual prognosis significance (p value: 0.0012~0.024). Moreover, the performance of the prognosis model has been benchmarked against the traditional LASSO and wavelet-based methods in the 3- and 5-year prediction AUC items, respectively. Specifically, the proposed model with discrete stationary wavelet base (bior1.5) overwhelmingly outperformed traditional LASSO and wavelet-based methods (AUC: 0.787 vs. 0.782 and 0.721 for the 3-year case; AUC: 0.937 vs. 0.802 and 0.859 for the 5-year case). Thus, the proposed model provides a more accurate perspective, but with less predictor burden for clinical prognosis in the pancreatic carcinoma study.

Video Recording and Analysis of Avian Movements and Behavior: Insights from Courtship Case Studies

Video recordings are useful tools for advancing our understanding of animal movements and behavior. Over the past decades, a burgeoning area of behavioral research has put forward innovative methods to investigate animal movement using video analysis, which includes motion capture and machine learning algorithms. These tools are particularly valuable for the study of elaborate and complex motor behaviors, but can be challenging to use. We focus in particular on elaborate courtship displays, which commonly involve rapid and/or subtle motor patterns. Here, we review currently available tools and provide hands-on guidelines for implementing these techniques in the study of avian model species. First, we suggest a set of possible strategies and solutions for video acquisition based on different model systems, environmental conditions, and time or financial budget. We then outline the available options for video analysis and illustrate how different analytical tools can be chosen to draw inference about animal motor performance. Finally, a detailed case study describes how these guidelines have been implemented to study courtship behavior in golden-collared manakins (Manacus vitellinus).

Potential surgical therapies for drug-resistant focal epilepsy

Drug-resistant focal epilepsy (DRFE), defined by failure of two antiepileptic drugs, affects 30% of epileptic patients. Epilepsy surgeries are alternative options for this population. Preoperative evaluation is critical to include potential candidates, and to choose the most appropriate procedure to maximize efficacy and simultaneously minimize side effects. Traditional procedures involve open skull surgeries and epileptic focus resection. Alternatively, neuromodulation surgeries use peripheral nerve or deep brain stimulation to reduce the activities of epileptogenic focus. With the advanced improvement of laser-induced thermal therapy (LITT) technique and its utilization in neurosurgery, magnetic resonance-guided LITT (MRgLITT) emerges as a minimal invasive approach for drug-resistant focal epilepsy. In the present review, we first introduce drug-resistant focal epilepsy and summarize the indications, pros and cons of traditional surgical procedures and neuromodulation procedures. And then, focusing on MRgLITT, we thoroughly discuss its history, its technical details, its safety issues, and current evidence on its clinical applications. A case report on MRgLITT is also included to illustrate the preoperational evaluation. We believe that MRgLITT is a promising approach in selected patients with drug-resistant focal epilepsy, although large prospective studies are required to evaluate its efficacy and side effects, as well as to implement a standardized protocol for its application.

Computer-Aided Intracranial EEG Signal Identification Method Based on a Multi-Branch Deep Learning Fusion Model and Clinical Validation

Surgical intervention or the control of drug-refractory epilepsy requires accurate analysis of invasive inspection intracranial EEG (iEEG) data. A multi-branch deep learning fusion model is proposed to identify epileptogenic signals from the epileptogenic area of the brain. The classical approach extracts multi-domain signal wave features to construct a time-series feature sequence and then abstracts it through the bi-directional long short-term memory attention machine (Bi-LSTM-AM) classifier. The deep learning approach uses raw time-series signals to build a one-dimensional convolutional neural network (1D-CNN) to achieve end-to-end deep feature extraction and signal detection. These two branches are integrated to obtain deep fusion features and results. Resampling is employed to split the imbalanced epileptogenic and non-epileptogenic samples into balanced subsets for clinical validation. The model is validated over two publicly available benchmark iEEG databases to verify its effectiveness on a private, large-scale, clinical stereo EEG database. The model achieves high sensitivity (97.78%), accuracy (97.60%), and specificity (97.42%) on the Bern–Barcelona database, surpassing the performance of existing state-of-the-art techniques. It is then demonstrated on a clinical dataset with an average intra-subject accuracy of 92.53% and cross-subject accuracy of 88.03%. The results suggest that the proposed method is a valuable and extremely robust approach to help researchers and clinicians develop an automated method to identify the source of iEEG signals.

A novel consciousness emotion recognition method using ERP components and MMSE

Objective.Electroencephalogram (EEG) based emotion recognition mainly extracts traditional features from time domain and frequency domain, and the classification accuracy is often low for the complex nature of EEG signals. However, to the best of our knowledge, the fusion of event-related potential (ERP) components and traditional features is not employed in emotion recognition, and the ERP components are only identified and analyzed by the psychology professionals, which is time-consuming and laborious.Approach.In order to recognize the consciousness and unconsciousness emotions, we propose a novel consciousness emotion recognition method using ERP components and modified multi-scale sample entropy (MMSE). Firstly, ERP components such as N200, P300 and N300 are automatically identified and extracted based on shapelet technique. Secondly, variational mode decomposition and wavelet packet decomposition are utilized to process EEG signals for obtaining different levels of emotional variational mode function (VMF), namelyVMFβ+γ, and then nonlinear feature MMSE of eachVMFβ+γare extracted. At last, ERP components and nonlinear feature MMSE are fused to generate a new feature vector, which is fed into random forest to classify the consciousness and unconsciousness emotions.Main results.Experimental results demonstrate that the average classification accuracy of our proposed method reach 94.42%, 94.88%, and 94.95% for happiness, horror and anger, respectively.Significance.Our study indicates that the fusion of ERP components and nonlinear feature MMSE is more effective for the consciousness and unconsciousness emotions recognition, which provides a new research direction and method for the study of nonlinear time series.

Detection of focal and non-focal EEG signals using non-linear features derived from empirical wavelet transform rhythms

Surgery is recommended for epilepsy diagnosis in cases where patients do not respond well to anti-epilepsy medications. Successful surgery is essentially dependent on the area suffered from epilepsy, i.e., focal area. Electroencephalogram (EEG) signals are considered a powerful tool to identify focal or non-focal (normal) areas. In this work, we propose an automated method for focal and non-focal EEG signal identification, taking into account non-linear features derived from rhythms in the empirical wavelet transform (EWT) domain. The research paradigm is related to the decomposition of EEG signals into the delta, theta, alpha, beta, and gamma rhythms through the development of the EWT. Specifically, various non-linear features are extracted from rhythms composed of Stein's unbiased risk estimation entropy, threshold entropy, centered correntropy, and information potential. From a statistical point of view, Kruskal-Wallis (KW) statistical test is then used to identify the significant features. The significant features obtained from the KW test are fed to support vector machine (SVM) and k-nearest neighbor (KNN) classifiers. The SURE entropy provides an average classification accuracy of 93% and 82.6% for small and entire datasets by utilizing SVM and KNN classifiers with a tenfold cross-validation method, respectively. It is observed that the proposed method is better and competitive in comparison with other studies for small and large data, respectively. The obtained outcome concludes that the proposed framework could be used for people with epilepsy and can help the physicians to validate the assessment.

Cognitive Processing Impacts High Frequency Intracranial EEG Activity of Human Hippocampus in Patients With Pharmacoresistant Focal Epilepsy

The electrophysiological EEG features such as high frequency oscillations, spikes and functional connectivity are often used for delineation of epileptogenic tissue and study of the normal function of the brain. The epileptogenic activity is also known to be suppressed by cognitive processing. However, differences between epileptic and healthy brain behavior during rest and task were not studied in detail. In this study we investigate the impact of cognitive processing on epileptogenic and non-epileptogenic hippocampus and the intracranial EEG features representing the underlying electrophysiological processes. We investigated intracranial EEG in 24 epileptic and 24 non-epileptic hippocampi in patients with intractable focal epilepsy during a resting state period and during performance of various cognitive tasks. We evaluated the behavior of features derived from high frequency oscillations, interictal epileptiform discharges and functional connectivity and their changes in relation to cognitive processing. Subsequently, we performed an analysis whether cognitive processing can contribute to classification of epileptic and non-epileptic hippocampus using a machine learning approach. The results show that cognitive processing suppresses epileptogenic activity in epileptic hippocampus while it causes a shift toward higher frequencies in non-epileptic hippocampus. Statistical analysis reveals significantly different electrophysiological reactions of epileptic and non-epileptic hippocampus during cognitive processing, which can be measured by high frequency oscillations, interictal epileptiform discharges and functional connectivity. The calculated features showed high classification potential for epileptic hippocampus (AUC = 0.93). In conclusion, the differences between epileptic and non-epileptic hippocampus during cognitive processing bring new insight in delineation between pathological and physiological processes. Analysis of computed iEEG features in rest and task condition can improve the functional mapping during pre-surgical evaluation and provide additional guidance for distinguishing between epileptic and non-epileptic structure which is absolutely crucial for achieving the best possible outcome with as little side effects as possible.

Detection of Focal and Non-Focal Electroencephalogram Signals Using Fast Walsh-Hadamard Transform and Artificial Neural Network

The discrimination of non-focal class (NFC) and focal class (FC), is vital in localizing the epileptogenic zone (EZ) during neurosurgery. In the conventional diagnosis method, the neurologist has to visually examine the long hour electroencephalogram (EEG) signals, which consumes time and is prone to error. Hence, in this present work, automated diagnosis of FC EEG signals from NFC EEG signals is developed using the Fast Walsh-Hadamard Transform (FWHT) method, entropies, and artificial neural network (ANN). The FWHT analyzes the EEG signals in the frequency domain and decomposes it into the Hadamard coefficients. Five different nonlinear features, namely approximate entropy (ApEn), log-energy entropy (LogEn), fuzzy entropy (FuzzyEn), sample entropy (SampEn), and permutation entropy (PermEn) are extracted from the decomposed Hadamard coefficients. The extracted features detail the nonlinearity in the NFC and the FC EEG signals. The judicious entropy features are supplied to the ANN classifier, with a 10-fold cross-validation method to classify the NFC and FC classes. Two publicly available datasets such as the University of Bonn and Bern-Barcelona dataset are used to evaluate the proposed approach. A maximum sensitivity of 99.70%, the accuracy of 99.50%, and specificity of 99.30% with the 3750 pairs of NFC and FC signal are achieved using the Bern-Barcelona dataset, while the accuracy of 92.80%, the sensitivity of 91%, and specificity of 94.60% is achieved using University of Bonn dataset. Compared to the existing technique, the proposed approach attained a maximum classification performance in both the dataset.

Cross-database evaluation of EEG based epileptic seizures detection driven by adaptive median feature baseline correction

OBJECTIVE

In long-term electroencephalogram (EEG) signals, automated classification of epileptic seizures is desirable in diagnosing epilepsy patients, as it otherwise depends on visual inspection. To the best of the author's knowledge, existing studies have validated their algorithms using cross-validation on the same database and less number of attempts have been made to extend their work on other databases to test the generalization capability of the developed algorithms. In this study, we present the algorithm for cross-database evaluation for classification of epileptic seizures using five EEG databases collected from different centers. The cross-database framework helps when sufficient epileptic seizures EEG data are not available to build automated seizure detection model.

METHODS

Two features, namely successive decomposition index and matrix determinant were extracted at a segmentation length of 4 s (50% overlap). Then, adaptive median feature baseline correction (AM-FBC) was applied to overcome the inter-patient and inter-database variation in the feature distribution. The classification was performed using a support vector machine classifier with leave-one-database-out cross-validation. Different classification scenarios were considered using AM-FBC, smoothing of the train and test data, and post-processing of the classifier output.

RESULTS

Simulation results revealed the highest area under the curve-sensitivity-specificity-false detections (per hour) of 1-1-1-0.15, 0.89-0.99-0.82-2.5, 0.99-0.73-1-1, 0.95-0.97-0.85-1.7, 0.99-0.99-0.92-1.1 using the Ramaiah Medical College and Hospitals, Children's Hospital Boston-Massachusetts Institute of Technology, Temple University Hospital, Maastricht University Medical Centre, and University of Bonn databases respectively.

CONCLUSIONS

We observe that the AM-FBC plays a significant role in improving seizure detection results by overcoming inter-database variation of feature distribution.

SIGNIFICANCE

To the best of the author's knowledge, this is the first study reporting on the cross-database evaluation of classification of epileptic seizures and proven to be better generalization capability when evaluated using five databases and can contribute to accurate and robust detection of epileptic seizures in real-time.

Deep Learning Approach for Epileptic Focus Localization

The task of epileptic focus localization receives great attention due to its role in an effective epileptic surgery. The clinicians highly depend on the intracranial EEG data to make a surgical decision related to epileptic subjects suffering from uncontrollable seizures. This surgery usually aims to remove the epileptogenic region which requires precise characterization of that area using the EEG recordings. In this paper, we propose two methods based on deep learning targeting accurate automatic epileptic focus localization using the non-stationary EEG recordings. Our first proposed method is based on semi-supervised learning, in which a deep convolutional autoencoder is trained and then the pre-trained encoder is used with multi-layer perceptron as a classifier. The goal is to determine the location of the EEG signal that is responsible for the epileptic activity. In the second proposed method, unsupervised learning scheme is implemented by merging deep convolutional variational autoencoder and K-means algorithm for clustering the iEEG signals into two distinct clusters based on the seizure source. The proposed methods automate and integrate the features extraction and classification processes instead of manually extracting the features as done in the previous studies. Dimensionality reduction is achieved using the autoencoder, while the important spatio-temporal features are extracted from the EEG recordings using the convolutional layers. Moreover, we implemented the inference network of the semi-supervised model on FPGA. The results of our experiments demonstrate high classification accuracy and clustering performance in localizing the epileptic focus compared with the state of the art.

Complexity analysis and dynamic characteristics of EEG using MODWT based entropies for identification of seizure onset

In this paper, complexity analysis and dynamic characteristics of electroencephalogram (EEG) signal based on maximal overlap discrete wavelet transform (MODWT) has been exploited for the identification of seizure onset. Since wavelet-based studies were well suited for classification of normal and epileptic seizure EEG, we have applied MODWT which is an improved version of discrete wavelet transform (DWT). The selection of optimal wavelet sub-band and features plays a crucial role to understand the brain dynamics in epileptic patients. Therefore, we have investigated MODWT using four different wavelets, namely Haar, Coif4, Dmey, and Sym4 sub-bands until seven levels. Further, we have explored the potentials of six entropies, namely sigmoid, Shannon, wavelet, Renyi, Tsallis, and Steins unbiased risk estimator (SURE) entropies in each sub-band. The sigmoid entropy extracted from Haar wavelet in sub-band D4 showed the highest accuracy of 98.44% using support vector machine classifier for the EEG collected from Ramaiah Medical College and Hospitals (RMCH). Further, the highest accuracy of 100% and 94.51% was achieved for the University of Bonn (UBonn) and CHB-MIT databases respectively. The findings of the study showed that Haar and Dmey wavelets were found to be computationally economical and expensive respectively. Besides, in terms of dynamic characteristics, MODWT results revealed that the highest energy present in sub-bands D2, D3, and D4 and entropies in those respective sub-bands outperformed other entropies in terms of classification results for RMCH database. Similarly, using all the entropies, sub-bands D5 and D6 outperformed other sub-bands for UBonn and CHB-MIT databases respectively. In conclusion, the comparison results of MODWT outperformed DWT.

Epileptic Focus Localization via Brain Network Analysis on Riemannian Manifolds

OBJECTIVE

Brain network connectivity analysis plays an important role in computer-aided automatic localization of seizure onset zone (SOZ) from Intracranial Electroencephalography (iEEG). However, how to accurately compute brain network dynamics is still not well addressed. This work aims to develop an effective measure to find out the dynamics for SOZ localization.

METHODS

Given multiple-channel iEEG signals, the ictal process involves continuous changes of information propagation. In each time slot, the connectivity relationship between channels can be represented as a matrix. Since the matrices from different time slots do not lie on vector spaces, the similarity between them cannot be computed directly. In this paper, we regard the matrices as points on a Riemannian manifold, so that the similarity can be measured by the geodesic distance on the manifold. It addresses the information-losing problem in existing methods using a vector to approximate a matrix. With the Riemannian method, the brain network dynamics are figured out by clustering methods. A temporal segmentation process is applied to refine the segments for SOZ localization.

RESULTS

Our method is evaluated on six epilepsy patients, and the SOZ localization performance is evaluated by the area under the curve (AUC) score. Overall, our method obtains an average AUC score of 0.875, which outperforms the existing approaches.

CONCLUSION

Our method preserves more information in measuring the relationship between brain connectivity descriptors, thus is more robust for SOZ localization.

SIGNIFICANCE

Our method has great potentials for clinical epilepsy treatments.

Grasshopper optimization algorithm-based approach for the optimization of ensemble classifier and feature selection to classify epileptic EEG signals

Epilepsy is one of the most common neurological disease worldwide. It is diagnosed by analyzing a long electroencephalogram (EEG) recording in a clinical environment, which may be much prone to errors and a time-consuming task. In this paper, a methodology for the classification of an epileptic seizure is proposed for analyzing EEG signals. EEG signal is decomposed into intrinsic mode functions (IMFs) using empirical mode decomposition (EMD). A fusion, of the extracted non-linear and spike-based features from each of the IMF signals, is made. The parameters of five machine learning algorithms; k-nearest neighbor (k-NN), extreme learning machine (ELM), random forest (RF), support vector machine (SVM), and artificial neural network (ANN) are optimized, as well as a set of the significant features is chosen using grasshopper optimization algorithm (GOA). These classifiers with their optimized parameters are ensembled together for the classification of epileptic seizures. The results show that ensemble classifier performs better than individual classifier. A comparison of the proposed methodology with state of the art epileptic seizure detection techniques is also made for validation. Graphical abstract ᅟ.

Change Point Detection for Piecewise Envelope Current Signal Based on Wavelet Transform

One of the key issues of the accurate parameters analysis for the piecewise envelope current signal is to position the change point precisely. Discrete wavelet transform (DWT) modulus maxima method can detect change point, but the detection window of DWT will cause suspicious change point. Besides, the amount of calculated data is very large in actual process of envelope current signal. Therefore, in this paper, the envelope is used instead of the original sampling data for DWT so as to reduce the calculation amount. What is more, combined with the sliding dislocation window method, the change point can be located accurately and the pseudo-change point can be eliminated. The simulation results as well as the electric locomotive current and forging machine current examples show that it is feasible to detect the change point precisely through the proposed method, which provides possibilities for real-time online monitoring of change point.