A Long Short-Term Memory neural network for the detection of epileptiform spikes and high frequency oscillations

A data augmentation procedure to improve detection of spike ripples in brain voltage recordings

Epilepsy is a major neurological disorder characterized by recurrent, spontaneous seizures. For patients with drug-resistant epilepsy, treatments include neurostimulation or surgical removal of the epileptogenic zone (EZ), the brain region responsible for seizure generation. Precise targeting of the EZ requires reliable biomarkers. Spike ripples - high-frequency oscillations that co-occur with large amplitude epileptic discharges - have gained prominence as a candidate biomarker. However, spike ripple detection remains a challenge. The gold-standard approach requires an expert manually visualize and interpret brain voltage recordings, which limits reproducibility and high-throughput analysis. Addressing these limitations requires more objective, efficient, and automated methods for spike ripple detection, including approaches that utilize deep neural networks. Despite advancements, dataset heterogeneity and scarcity severely limit machine learning performance. Our study explores long-short term memory (LSTM) neural network architectures for spike ripple detection, leveraging data augmentation to improve classifier performance. We highlight the potential of combining training on augmented and in vivo data for enhanced spike ripple detection and ultimately improving diagnostic accuracy in epilepsy treatment.

Discovering Neurophysiological Characteristics of Pathological High-Frequency Oscillations in Epilepsy with an Explainable Deep Generative Model

Interictal high-frequency oscillation (HFO) is a promising biomarker of the epileptogenic zone (EZ). However, objective definitions to distinguish between pathological and physiological HFOs have remained elusive, impeding HFOs' clinical applications. We employed self-supervised deep generative variational autoencoders to learn such discriminative HFO features directly from their morphologies in a data-driven manner. We studied a large retrospective cohort of 185 patients who underwent intracranial monitoring and analyzed 686,410 candidate HFO events collected from 18,265 brain contacts across diverse brain regions. The model automatically clustered HFOs into distinct morphological groups in the latent space. One cluster consisted of putative morphologically defined pathological HFOs (mpHFOs): HFOs in that cluster were observed to be associated with spikes and exhibited high signal intensity both in the HFO band (>80 Hz) at detection and in the sub-HFO band (10-80 Hz) surrounding the detection and were primarily localized in the seizure onset zone (SOZ). Moreover, resection of brain regions based on a higher prevalence of interictal mpHFOs better predicted postoperative seizure outcomes than current clinical standards based on SOZ removal. Our self-supervised, explainable, deep generative model distills pathological HFOs and thus potentially helps delineate the EZ purely from interictal intracranial EEG data.

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.

The detection of absence seizures using cross-frequency coupling analysis with a deep learning network

High frequency oscillations are important novel biomarkers of epileptogenic tissue. The interaction of oscillations across different time scales is revealed as cross-frequency coupling (CFC) representing a high-order structure in the functional organization of brain rhythms. New artificial intelligence methods such as deep learning neural networks can provide powerful tools for automated analysis of EEG. Here we present a Stacked Sparse Autoencoder (SSAE) trained to recognize absence seizure activity based on the cross-frequency patterns within scalp EEG. We used EEG records from the Temple University Hospital database. Absence seizures (n = 94) from 12 patients were taken into analysis along with segments of background activity. Half of the records were selected randomly for network training and the second half were used for testing. Power-to-power coupling was calculated between all frequencies 2-120 Hz pairwise using the EEGLAB toolbox. The resulting CFC matrices were used as training or testing inputs to the autoencoder. The trained network was able to recognize background and seizure segments (not used in training) with a sensitivity of 96.3%, specificity of 99.8% and overall accuracy of 98.5%. Our results provide evidence that the SSAE neural networks can be used for automated detection of absence seizures within scalp EEG.

V2IED: Dual-view learning framework for detecting events of interictal epileptiform discharges

Interictal epileptiform discharges (IED) as large intermittent electrophysiological events are associated with various severe brain disorders. Automated IED detection has long been a challenging task, and mainstream methods largely focus on singling out IEDs from backgrounds from the perspective of waveform, leaving normal sharp transients/artifacts with similar waveforms almost unattended. An open issue still remains to accurately detect IED events that directly reflect the abnormalities in brain electrophysiological activities, minimizing the interference from irrelevant sharp transients with similar waveforms only. This study then proposes a dual-view learning framework (namely V2IED) to detect IED events from multi-channel EEG via aggregating features from the two phases: (1) Morphological Feature Learning: directly treating the EEG as a sequence with multiple channels, a 1D-CNN (Convolutional Neural Network) is applied to explicitly learning the deep morphological features; and (2) Spatial Feature Learning: viewing the EEG as a 3D tensor embedding channel topology, a CNN captures the spatial features at each sampling point followed by an LSTM (Long Short-Term Memories) to learn the evolution of these features. Experimental results from a public EEG dataset against the state-of-the-art counterparts indicate that: (1) compared with the existing optimal models, V2IED achieves a larger area under the receiver operating characteristic (ROC) curve in detecting IEDs from normal sharp transients with a 5.25% improvement in accuracy; (2) the introduction of spatial features improves performance by 2.4% in accuracy; and (3) V2IED also performs excellently in distinguishing IEDs from background signals especially benign variants.

A review of signal processing and machine learning techniques for interictal epileptiform discharge detection

Brain interictal epileptiform discharges (IEDs), as one of the hallmarks of epileptic brain, are transient events captured by electroencephalogram (EEG). IEDs are generated by seizure networks, and they occur between seizures (interictal periods). The development of a robust method for IED detection could be highly informative for clinical treatment procedures and epileptic patient management. Since 1972, different machine learning techniques, from template matching to deep learning, have been developed to automatically detect IEDs from scalp EEG (scEEG) and intracranial EEG (iEEG). While the scEEG signals suffer from low information details and high attenuation of IEDs due to the high skull electrical impedance, the iEEG signals recorded using implanted electrodes enjoy higher details and are more suitable for identifying the IEDs. In this review paper, we group IED detection techniques into six categories: (1) template matching, (2) feature representation (mimetic, time-frequency, and nonlinear features), (3) matrix decomposition, (4) tensor factorization, (5) neural networks, and (6) estimation of the iEEG from the concurrent scEEG followed by detection and classification. The methods are compared quantitatively (e.g., in terms of accuracy, sensitivity, and specificity), and their general advantages and limitations are described. Finally, current limitations and possible future research paths related to this field are mentioned.

Application of bidirectional long short-term memory network for prediction of cognitive age

Electroencephalography (EEG) measures changes in neuronal activity and can reveal significant changes from infancy to adulthood concomitant with brain maturation, making it a potential physiological marker of brain maturation and cognition. To investigate a promising deep learning tool for EEG classification, we applied the bidirectional long short-term memory (BLSTM) algorithm to analyze EEG data from the pediatric EEG laboratory of Taipei Tzu Chi Hospital. The trained BLSTM model was 86% accurate when identifying EEGs from young children (8 months-6 years) and adolescents (12-20 years). However, there was only a modest classification accuracy (69.3%) when categorizing EEG samples into three age groups (8 months-6 years, 6-12 years, and 12-20 years). For EEG samples from patients with intellectual disability, the prediction accuracy of the trained BLSTM model was 46.4%, which was significantly lower than its accuracy for EEGs from neurotypical patients, indicating that the individual's intelligence plays a major role in the age prediction. This study confirmed that scalp EEG can reflect brain maturation and the BLSTM algorithm is a feasible deep learning tool for the identification of cognitive age. The trained model can potentially be applied to clinical services as a supportive measurement of neurodevelopmental status.

Graph neural networks in EEG spike detection

OBJECTIVE

This study develops new machine learning architectures that are more adept at detecting interictal epileptiform discharges (IEDs) in scalp EEG. A comparison of results using the average precision (AP) metric is made with the proposed models on two datasets obtained from Baptist Hospital of Miami and Temple University Hospital.

METHODS

Applying graph neural networks (GNNs) on functional connectivity (FC) maps of different frequency sub-bands to yield a novel architecture we call FC-GNN. Attention mechanism is applied on a complete graph to let the neural network select its important edges, hence bypassing the extraction of features, a model we refer to as CA-GNN.

RESULTS

On the Baptist Hospital dataset, the results were as follows: Vanilla Self-Attention →0.9029±0.0431, Hierarchical Attention →0.8546±0.0587, Vanilla Visual Geometry Group (VGG) →0.92±0.0618, Satelight →0.9219±0.046, FC-GNN →0.9731±0.0187, and CA-GNN →0.9788±0.0125. In the same order, the results on the Temple University Hospital dataset are 0.9692, 0.9113, 0.97, 0.9575, 0.963, and 0.9879.

CONCLUSION

Based on the good results they yield, GNNs prove to have a strong potential in detecting epileptogenic activity.

SIGNIFICANCE

This study opens the door for the discovery of the powerful role played by GNNs in capturing IEDs, which is an essential step for identifying the epileptogenic networks of the affected brain and hence improving the prospects for more accurate 3D source localization.

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.

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.

Recurrent neural network to predict hyperelastic constitutive behaviors of the skeletal muscle

Hyperelastic constitutive laws have been commonly used to model the passive behavior of the human skeletal muscle. Despite many efforts, the use of accurate finite element formulations of hyperelastic constitutive laws is still time-consuming for a real-time medical simulation system. The objective of the present study was to develop a deep learning model to predict the hyperelastic constitutive behaviors of the skeletal muscle toward a fast estimation of the muscle tissue stress.A finite element (FE) model of the right psoas muscle was developed. Neo-Hookean and Mooney-Rivlin laws were used. A tensile test was performed with an applied body force. A learning database was built from this model using an automatic probabilistic generation process. A long-short term memory (LSTM) neural network was implemented to predict the stress evolution of the skeletal muscle tissue. A hyperparameter tuning process was conducted. Root mean square error (RMSE) and associated relative error was quantified to evaluate the precision of the predictive capacity of the developed deep learning model. Pearson correlation coefficients (R) was also computed.The nodal displacements and the maximal stresses range from 70 to 227 mm and from 2.79 to 5.61 MPa for Neo-Hookean and Monney-Rivlin laws, respectively. Regarding the LSTM predictions, the RMSE ranges from 224.3 ± 3.9 Pa (8%) to 227.5 [Formula: see text] 5.7 Pa (4%) for Neo-Hookean and Monney-Rivlin laws, respectively. Pearson correlation coefficients (R) of 0.78 [Formula: see text] 0.02 and 0.77 [Formula: see text] 0.02 were obtained for Neo-Hookean and Monney-Rivlin laws, respectively.The present study showed that, for the first time, the use of a deep learning model can reproduce the time-series behaviors of the complex FE formulations for skeletal muscle modeling. In particular, the use of a LSTM neural network leads to a fast and accurate surrogate model for the in silico prediction of the hyperelastic constitutive behaviors of the skeletal muscle. As perspectives, the developed deep learning model will be integrated into a real-time medical simulation of the skeletal muscle for prosthetic socket design and childbirth simulator.

Engineering nonlinear epileptic biomarkers using deep learning and Benford's law

In this study, we designed two deep neural networks to encode 16 features for early seizure detection in intracranial EEG and compared them and their frequency responses to 16 widely used engineered metrics to interpret their properties: epileptogenicity index (EI), phase locked high gamma (PLHG), time and frequency domain Cho Gaines distance (TDCG, FDCG), relative band powers, and log absolute band powers (from alpha, beta, theta, delta, low gamma, and high gamma bands). The deep learning models were pretrained for seizure identification on the time and frequency domains of 1 s, single-channel clips of 127 seizures (from 25 different subjects) using "leave-one-out" (LOO) cross validation. Each neural network extracted unique feature spaces that were interpreted using spectral power modulations before being used to train a Random Forest Classifier (RFC) for seizure identification. The Gini Importance of each feature was calculated from the pretrained RFC, enabling the most significant features (MSFs) for each task to be identified. The MSFs were extracted to train another RFC for UPenn and Mayo Clinic's Seizure Detection Kaggle Challenge. They obtained an AUC score of 0.93, demonstrating a transferable method to identify and interpret biomarkers for seizure detection.

Higher-order tensor decomposition based scalp-to-intracranial EEG projection for detection of interictal epileptiform discharges

Objective.Interictal epileptiform discharges (IEDs) occur between two seizures onsets. IEDs are mainly captured by intracranial recordings and are often invisible over the scalp. This study proposes a model based on tensor factorization to map the time-frequency (TF) features of scalp EEG (sEEG) to the TF features of intracranial EEG (iEEG) in order to detect IEDs from over the scalp with high sensitivity.Approach.Continuous wavelet transform is employed to extract the TF features. Time, frequency, and channel modes of IED segments from iEEG recordings are concatenated into a four-way tensor. Tucker and CANDECOMP/PARAFAC decomposition techniques are employed to decompose the tensor into temporal, spectral, spatial, and segmental factors. Finally, TF features of both IED and non-IED segments from scalp recordings are projected onto the temporal components for classification.Main results.The model performance is obtained in two different approaches: within- and between-subject classification approaches. Our proposed method is compared with four other methods, namely a tensor-based spatial component analysis method, TF-based method, linear regression mapping model, and asymmetric-symmetric autoencoder mapping model followed by convolutional neural networks. Our proposed method outperforms all these methods in both within- and between-subject classification approaches by respectively achieving 84.2% and 72.6% accuracy values.Significance.The findings show that mapping sEEG to iEEG improves the performance of the scalp-based IED detection model. Furthermore, the tensor-based mapping model outperforms the autoencoder- and regression-based mapping models.

AiED: Artificial intelligence for the detection of intracranial interictal epileptiform discharges

OBJECTIVE

Deep learning provides an appealing solution for the ongoing challenge of automatically classifying intracranial interictal epileptiform discharges (IEDs). We report results from an automated method consisting of a template-matching algorithm and convolutional neural network (CNN) for the detection of intracranial IEDs ("AiED").

METHODS

1000 intracranial electroencephalogram (EEG) epochs extracted randomly from 307 subjects with refractory epilepsy were annotated independently by two expert neurophysiologists. These annotated epochs were divided into 1062 two-second epochs with IEDs and 1428 two-second epochs without IEDs, which were transformed into spectrograms prior to training the neural network. The highest performing network was validated on an annotated external test set.

RESULTS

The final network had an F1-score of 0.95 (95% CI: 0.91-0.98) and an average Area Under the Receiver Operating Characteristic of 0.98 (95% CI: 0.96-1.00). For the external test set, it showed an overall F1-score of 0.71, correctly identifying 100% of all high-amplitude IED complexes, 96.23% of all high-amplitude isolated IEDs, and 66.15% of all IEDs of atypical morphology.

CONCLUSIONS

Template-matching combined with a CNN offers a fast, robust method for detecting intracranial IEDs.

SIGNIFICANCE

"AiED" is generalizable and achieves comparable performance to human reviewers; it may support clinical and research EEG analyses.

A channel independent generalized seizure detection method for pediatric epileptic seizures

BACKGROUND AND OBJECTIVE

Epilepsy the disorder of the central nervous system has its worldwide presence in roughly 50 million people as estimated by the World Health Organization. Electroencephalogram (EEG) is one of the most common and non-invasive ways of analyzing and studying the subtle changes in neuronal activity of the brain during an epileptic seizure attack. These changes can be analyzed for developing an automated system that would assert the chances of an impending seizure. As changeable nature of seizure affects the patients from having a normal life, hence progress in developing new methods will improve the quality of life and also provide assistance in the medical sector. Objective of the proposed method is to avoid EEG channel selection and use all input EEG channel features to design a generalized epileptic seizure detection framework.

METHOD

In this work, a long short-term memory network has been proposed that is not complex and has the capability of effectively detecting epileptic seizures from both non-invasive and invasive electroencephalogram recordings. The proposed framework is simple and effective and designed in such capacity that raw electroencephalogram signals can be used to detect seizures. Also, a generalized approach has been followed that is channel independent such that EEG signals belonging to any hemisphere of the brain can be detected effectively by the proposed architecture.

RESULTS

The automated seizure detection system achieved high seizure detection sensitivity of 99.9%, and a low false-positive rate of 0.003 per hour for the Children's Hospital Boston-Massachusetts Institute of Technology dataset. While for the Sleep-Wake-Epilepsy-Center of the University Department of Neurology at the Inselspital Bern dataset, the sensitivity is 99.4% and false-positive rate of 0.006 per hour. Convergence analysis of the proposed model provides a significant amount of reliability and correctness in the efficient detection of epileptic seizures.

CONCLUSION

Assessment of the proposed framework on non-invasive as well as invasive EEG signals showed that the framework worked well for different type of EEG recordings as different metrics gave satisfactory results. As the framework is simple and did not require any additional parameter optimization techniques, it reduced the processing overheads without affecting the accuracy. Hence, it can be used as an efficient method for monitoring epileptic seizures.

A Novel Approach to Learning Models on EEG Data Using Graph Theory Features—A Comparative Study

Brain connectivity is studied as a functionally connected network using statistical methods such as measuring correlation or covariance. The non-invasive neuroimaging techniques such as Electroencephalography (EEG) signals are converted to networks by transforming the signals into a Correlation Matrix and analyzing the resulting networks. Here, four learning models, namely, Logistic Regression, Random Forest, Support Vector Machine, and Recurrent Neural Networks (RNN), are implemented on two different types of correlation matrices: Correlation Matrix (static connectivity) and Time-resolved Correlation Matrix (dynamic connectivity), to classify them either on their psychometric assessment or the effect of therapy. These correlation matrices are different from traditional learning techniques in the sense that they incorporate theory-based graph features into the learning models, thus providing novelty to this study. The EEG data used in this study is trail-based/event-related from five different experimental paradigms, of which can be broadly classified as working memory tasks and assessment of emotional states (depression, anxiety, and stress). The classifications based on RNN provided higher accuracy (74–88%) than the other three models (50–78%). Instead of using individual graph features, a Correlation Matrix provides an initial test of the data. When compared with the Time-resolved Correlation Matrix, it offered a 4–5% higher accuracy. The Time-resolved Correlation Matrix is better suited for dynamic studies here; it provides lower accuracy when compared to the Correlation Matrix, a static feature.

BECT Spike Detection Based on Novel EEG Sequence Features and LSTM Algorithms

The benign epilepsy with spinous waves in the central temporal region (BECT) is the one of the most common epileptic syndromes in children, that seriously threaten the nervous system development of children. The most obvious feature of BECT is the existence of a large number of electroencephalogram (EEG) spikes in the Rolandic area during the interictal period, that is an important basis to assist neurologists in BECT diagnosis. With this regard, the paper proposes a novel BECT spike detection algorithm based on time domain EEG sequence features and the long short-term memory (LSTM) neural network. Three time domain sequence features, that can obviously characterize the spikes of BECT, are extracted for EEG representation. The synthetic minority oversampling technique (SMOTE) is applied to address the spike imbalance issue in EEGs, and the bi-directional LSTM (BiLSTM) is trained for spike detection. The algorithm is evaluated using the EEG data of 15 BECT patients recorded from the Children's Hospital, Zhejiang University School of Medicine (CHZU). The experiment shows that the proposed algorithm can obtained an average of 88.54% F1 score, 92.04% sensitivity, and 85.75% precision, that generally outperforms several state-of-the-art spike detection methods.

An integrated deep learning model for motor intention recognition of multi-class EEG Signals in upper limb amputees

BACKGROUND AND OBJECTIVE

Recognition of motor intention based on electroencephalogram (EEG) signals has attracted considerable research interest in the field of pattern recognition due to its notable application of non-muscular communication and control for those with severe motor disabilities. In analysis of EEG data, achieving a higher classification performance is dependent on the appropriate representation of EEG features which is mostly characterized by one unique frequency before applying a learning model. Neglecting other frequencies of EEG signals could deteriorate the recognition performance of the model because each frequency has its unique advantages. Motivated by this idea, we propose to obtain distinguishable features with different frequencies by introducing an integrated deep learning model to accurately classify multiple classes of upper limb movement intentions.

METHODS

The proposed model is a combination of long short-term memory (LSTM) and stacked autoencoder (SAE). To validate the method, four high-level amputees were recruited to perform five motor intention tasks. The acquired EEG signals were first preprocessed before exploring the consequence of input representation on the performance of LSTM-SAE by feeding four frequency bands related to the tasks into the model. The learning model was further improved by t-distributed stochastic neighbor embedding (t-SNE) to eliminate feature redundancy, and to enhance the motor intention recognition.

RESULTS

The experimental results of the classification performance showed that the proposed model achieves an average performance of 99.01% for accuracy, 99.10% for precision, 99.09% for recall, 99.09% for f1_score, 99.77% for specificity, and 99.0% for Cohen's kappa, across multi-subject and multi-class scenarios. Further evaluation with 2-dimensional t-SNE revealed that the signal decomposition has a distinct multi-class separability in the feature space.

CONCLUSION

This study demonstrated the predominance of the proposed model in its ability to accurately classify upper limb movements from multiple classes of EEG signals, and its potential application in the development of a more intuitive and naturalistic prosthetic control.

Deep learning for robust detection of interictal epileptiform discharges

OBJECTIVE

Automatic detection of interictal epileptiform discharges (IEDs, short as ``spikes'') from an epileptic brain can help predict seizure recurrence and support the diagnosis of epilepsy. Developing fast, reliable and robust detection methods for IEDs based on scalp or intracortical EEG may facilitate online seizure monitoring and closed-loop neurostimulation.

APPROACH

We developed a new deep learning approach, which employs a long short-term memory (LSTM) network architecture (``IEDnet'') and an auxiliary classifier generative adversarial network (AC-GAN), to train on both expert-annotated and augmented spike events from intracranial electroencephalography (iEEG) recordings of epilepsy patients. We validated our IEDnet with two real-world iEEG datasets, and compared IEDnet with the support vector machine (SVM) and random forest (RF) classifiers on their detection performances.

MAIN RESULTS

IEDnet achieved excellent cross-validated detection performances in terms of both sensitivity and specificity, and outperformed SVM and RF. Synthetic spike samples augmented by AC-GAN further improved the detection performance. In addition, the performance of IEDnet was robust with respect to the sampling frequency and noise. Furthermore, we also demonstrated the cross-institutional generalization ability of IEDnet while testing between two datasets.

SIGNIFICANCE

IEDnet achieves excellent detection performances in identifying interictal spikes. AC-GAN can produce augmented iEEG samples to improve supervised deep learning.

An Automatic HFO Detection Method Combining Visual Inspection Features with Multi-Domain Features

As an important promising biomarker, high frequency oscillations (HFOs) can be used to track epileptic activity and localize epileptogenic zones. However, visual marking of HFOs from a large amount of intracranial electroencephalogram (iEEG) data requires a great deal of time and effort from researchers, and is also very dependent on visual features and easily influenced by subjective factors. Therefore, we proposed an automatic epileptic HFO detection method based on visual features and non-intuitive multi-domain features. To eliminate the interference of continuous oscillatory activity in detected sporadic short HFO events, the iEEG signals adjacent to the detected events were set as the neighboring environmental range while the number of oscillations and the peak-valley differences were calculated as the environmental reference features. The proposed method was developed as a MatLab-based HFO detector to automatically detect HFOs in multi-channel, long-distance iEEG signals. The performance of our detector was evaluated on iEEG recordings from epileptic mice and patients with intractable epilepsy. More than 90% of the HFO events detected by this method were confirmed by experts, while the average missed-detection rate was < 10%. Compared with recent related research, the proposed method achieved a synchronous improvement of sensitivity and specificity, and a balance between low false-alarm rate and high detection rate. Detection results demonstrated that the proposed method performs well in sensitivity, specificity, and precision. As an auxiliary tool, our detector can greatly improve the efficiency of clinical experts in inspecting HFO events during the diagnosis and treatment of epilepsy.

RippleNet: a Recurrent Neural Network for Sharp Wave Ripple (SPW-R) Detection

Hippocampal sharp wave ripples (SPW-R) have been identified as key bio-markers of important brain functions such as memory consolidation and decision making. Understanding their underlying mechanisms in healthy and pathological brain function and behaviour rely on accurate SPW-R detection. In this multidisciplinary study, we propose a novel, self-improving artificial intelligence (AI) detection method in the form of deep Recurrent Neural Networks (RNN) with Long Short-Term memory (LSTM) layers that can learn features of SPW-R events from raw, labeled input data. The approach contrasts conventional routines that typically relies on hand-crafted, heuristic feature extraction and often laborious manual curation. The algorithm is trained using supervised learning on hand-curated data sets with SPW-R events obtained under controlled conditions. The input to the algorithm is the local field potential (LFP), the low-frequency part of extracellularly recorded electric potentials from the CA1 region of the hippocampus. Its output predictions can be interpreted as time-varying probabilities of SPW-R events for the duration of the inputs. A simple thresholding applied to the output probabilities is found to identify times of SPW-R events with high precision. The non-causal, or bidirectional variant of the proposed algorithm demonstrates consistently better accuracy compared to the causal, or unidirectional counterpart. Reference implementations of the algorithm, named ‘RippleNet’, are open source, freely available, and implemented using a common open-source framework for neural networks (tensorflow.keras) and can be easily incorporated into existing data analysis workflows for processing experimental data.

Statistical Features in High-Frequency Bands of Interictal iEEG Work Efficiently in Identifying the Seizure Onset Zone in Patients with Focal Epilepsy

The design of a computer-aided system for identifying the seizure onset zone (SOZ) from interictal and ictal electroencephalograms (EEGs) is desired by epileptologists. This study aims to introduce the statistical features of high-frequency components (HFCs) in interictal intracranial electroencephalograms (iEEGs) to identify the possible seizure onset zone (SOZ) channels. It is known that the activity of HFCs in interictal iEEGs, including ripple and fast ripple bands, is associated with epileptic seizures. This paper proposes to decompose multi-channel interictal iEEG signals into a number of subbands. For every 20 s segment, twelve features are computed from each subband. A mutual information (MI)-based method with grid search was applied to select the most prominent bands and features. A gradient-boosting decision tree-based algorithm called LightGBM was used to score each segment of the channels and these were averaged together to achieve a final score for each channel. The possible SOZ channels were localized based on the higher value channels. The experimental results with eleven epilepsy patients were tested to observe the efficiency of the proposed design compared to the state-of-the-art methods.

Supervised machine learning tools: a tutorial for clinicians

In an increasingly data-driven world, artificial intelligence is expected to be a key tool for converting big data into tangible benefits and the healthcare domain is no exception to this. Machine learning aims to identify complex patterns in multi-dimensional data and use these uncovered patterns to classify new unseen cases or make data-driven predictions. In recent years, deep neural networks have shown to be capable of producing results that considerably exceed those of conventional machine learning methods for various classification and regression tasks. In this paper, we provide an accessible tutorial of the most important supervised machine learning concepts and methods, including deep learning, which are potentially the most relevant for the medical domain. We aim to take some of the mystery out of machine learning and depict how machine learning models can be useful for medical applications. Finally, this tutorial provides a few practical suggestions for how to properly design a machine learning model for a generic medical problem.

Variability analysis of epileptic EEG using the maximal overlap discrete wavelet transform

Purpose

To determine if there is a difference in the wavelet variances of seizure and non-seizure channels in the EEG of an epileptic subject.

Methods

A six-level decomposition was applied using the Maximal Overlap Discrete Wavelet Transform (MODWT). The wavelet variance and 95% CIs were calculated for each level of the decomposition. The number of changes in variance for each level were found using a change-point detection method of Whitcher. The Kruskal-Wallis test was used to determine if there were differences in the median number of change points within channels and across frequency bands (levels).

Results

No distinctive pattern was found for the wavelet variances to differentiate the seizure and non-seizure channels. The seizure channels tended to have lower variances for each level and overall, but this pattern only held for one of the three seizure channels (RAST4). The median number of change points did not differ between the seizure and non-seizure channels either within each channel or across the frequency bands.

Conclusion

The use of the MODWT in examining the variances and changes in variance did not show specific patterns which differentiate between seizure and non-seizure channels.

Modeling Volatility Characteristics of Epileptic EEGs using GARCH Models

Objective: To determine if there was a difference in the volatility characteristics of seizure and non-seizure onset channels in the intracranial electroencephalogram (EEG) in a patient with temporal lobe epilepsy. Methods: The half-life of volatility for the different EEG channels was determined using Autoregressive Moving Average–Generalized Autoregressive Conditional Heteroscedasticity (ARMA–GARCH) models; confidence intervals were constructed using the delta method and an asymptotic method for comparing the half-lives. Results: Clinically determined seizure onsets occurred over strip electrodes named RAST (Right Anterior Subtemporal) and RMST (Right Mid Subtemporal), at locations 2, 3 and 4, on the strip electrodes. The half-lives of volatility for two of the three seizure channels, RAST3 and RAST4, were found to be significantly lower the rest of the channels for six one-minute EEG segments prior to seizure onset and nine one-minute EEG segments of an awake state. The half-lives of volatility for RAST3 and RAST4 were not significantly different to the non-seizure channels for ten one-minute segments of sleep and ten one-minute segments of sleep-to-awake states. The estimates for the half-lives were consistent for randomly selected one-minute EEG segments. Conclusions: The use of GARCH models may be a useful tool in determining hidden properties in epileptiform EEGs that may lead to better understanding of the seizure generating process.

Multiband entropy-based feature-extraction method for automatic identification of epileptic focus based on high-frequency components in interictal iEEG

Presurgical investigations for categorizing focal patterns are crucial, leading to localization and surgical removal of the epileptic focus. This paper presents a machine learning approach using information theoretic features extracted from high-frequency subbands to detect the epileptic focus from interictal intracranial electroencephalogram (iEEG). It is known that high-frequency subbands (>80 Hz) include important biomarkers such as high-frequency oscillations (HFOs) for identifying epileptic focus commonly referred to as the seizure onset zone (SOZ). In this analysis, the multi-channel interictal iEEG signals were splitted into segments and each segment was decomposed into multiple high-frequency subbands. The different types of entropy were calculated for each of the subbands and the sparse linear discriminant analysis (sLDA) was applied to select the prominent entropy features. Due to the imbalance of SOZ and non-SOZ channels in iEEG data, the use of machine learning techniques is always tricky. To deal with the imbalanced learning problem, an adaptive synthetic oversampling approach (ADASYN) with radial basis function kernel-based SVM was used to detect the focal segments. Finally, the epileptic focus was identified based on detection of focal segments on SOZ and non-SOZ channels. Eight patients were examined to observe the efficiency of the automatic detector. The experimental results and statistical tests indicate that the proposed automatic detector can identify the epileptic focus accurately and efficiently.