Epileptic Seizure Prediction Using Spatiotemporal Feature Fusion on EEG

Electroencephalography (EEG) plays a crucial role in epilepsy analysis, and epileptic seizure prediction has significant value for clinical treatment of epilepsy. Currently, prediction methods using Convolutional Neural Network (CNN) primarily focus on local features of EEG, making it challenging to simultaneously capture the spatial and temporal features from multi-channel EEGs to identify the preictal state effectively. In order to extract inherent spatial relationships among multi-channel EEGs while obtaining their temporal correlations, this study proposed an end-to-end model for the prediction of epileptic seizures by incorporating Graph Attention Network (GAT) and Temporal Convolutional Network (TCN). Low-pass filtered EEG signals were fed into the GAT module for EEG spatial feature extraction, and followed by TCN to capture temporal features, allowing the end-to-end model to acquire the spatiotemporal correlations of multi-channel EEGs. The system was evaluated on the publicly available CHB-MIT database, yielding segment-based accuracy of 98.71%, specificity of 98.35%, sensitivity of 99.07%, and F1-score of 98.71%, respectively. Event-based sensitivity of 97.03% and False Positive Rate (FPR) of 0.03/h was also achieved. Experimental results demonstrated this system can achieve superior performance for seizure prediction by leveraging the fusion of EEG spatiotemporal features without the need of feature engineering.

Combining temporal and spatial attention for seizure prediction

Purpose

Approximately 1% of the world population is currently suffering from epilepsy. Successful seizure prediction is necessary for those patients. Influenced by neurons in their own and surrounding locations, the electroencephalogram (EEG) signals collected by scalp electrodes carry information of spatiotemporal interactions. Therefore, it is a great challenge to exploit the spatiotemporal information of EEG signals fully.

Methods

In this paper, a new seizure prediction model called Gatformer is proposed by fusing the graph attention network (GAT) and the Transformer. The temporal and spatial attention are combined to extract EEG information from the perspective of spatiotemporal interactions. The model aims to explore the temporal dependence of single-channel EEG signals and the spatial correlations among multi-channel EEG signals. It can automatically identify the most noteworthy interaction in brain regions and achieve accurate seizure prediction.

Results

Compared with the baseline models, the performance of our model is significantly improved. The false prediction rate (FPR) on the private dataset is 0.0064/h. The average accuracy, specificity and sensitivity are 98.25%, 99.36% and 97.65%.

Conclusion

The proposed model is comparable to the state of the arts. Experiments on different datasets show that it has good robustness and generalization performance. The high sensitivity and low FPR prove that this model has great potential to realize clinical assistance for diagnosis and treatment.

Diverse nature of interictal oscillations: EEG-based biomarkers in epilepsy

Interictal electroencephalogram (EEG) patterns, including high-frequency oscillations (HFOs), interictal spikes (ISs), and slow wave activities (SWAs), are defined as specific oscillations between seizure events. These interictal oscillations reflect specific dynamic changes in network excitability and play various roles in epilepsy. In this review, we briefly describe the electrographic characteristics of HFOs, ISs, and SWAs in the interictal state, and discuss the underlying cellular and network mechanisms. We also summarize representative evidence from experimental and clinical epilepsy to address their critical roles in ictogenesis and epileptogenesis, indicating their potential as electrophysiological biomarkers of epilepsy. Importantly, we put forwards some perspectives for further research in the field.

Generalizable epileptic seizures prediction based on deep transfer learning

Predicting seizures before they happen can help prevent them through medication. In this research, first, a total of 22 features were extracted from 5-s segmented EEG signals. Second, tensors were developed as inputs for different deep transfer learning models to find the best model for predicting epileptic seizures. The effect of Pre-ictal state duration was also investigated by selecting four different intervals of 10, 20, 30, and 40 min. Then, nine models were created by combining three ImageNet convolutional networks with three classifiers and were examined for predicting seizures patient-dependently. The Xception convolutional network with a Fully Connected (FC) classifier achieved an average sensitivity of 98.47% and a False Prediction Rate (FPR) of 0.031 h-1 in a 40-min Pre-ictal state for ten patients from the European database. The most promising result of this study was the patient-independent prediction of epileptic seizures; the MobileNet-V2 model with an FC classifier was trained with one patient's data and tested on six other patients, achieving a sensitivity rate of 98.39% and an FPR of 0.029 h-1 for a 40-min Pre-ictal scheme.

Machine Learning for Bioelectronics on Wearable and Implantable Devices: Challenges and Potential

Bioelectronics presents a promising future in the field of embedded and implantable electronics, providing a range of functional applications, from personal health monitoring to bioactuators. However, due to the intrinsic difficulties present in producing and optimizing bioelectronics, recent research has focused on utilizing machine learning (ML) to reliably mitigate such issues and aid in process development. This review focuses on the recent developments of integrating ML into bioelectronics, aiding in a multitude of areas, such as material development, fabrication process optimization, and system integration. First, discussing how ML has aided in the material development by identifying complex relationships between process input parameters and desired outputs, such as product design. Second, examine the advancements in ML to accurately optimize fabrication precision and stability for various 3D printing technologies. Third, provide an overview of how ML can greatly assist in the analysis of complex, nonlinear relationships in data obtained from bioelectronics. Lastly, a summary of the challenges present with utilizing ML with bioelectronics and any other developments in this field. Such advancements in the field of bioelectronics and ML could hopefully build a strong foundation for this research field, promoting smart optimization together with effective use of ML to further enhance the effectiveness of such applications. Impact statement The article serves to give insight about the use of the machine learning (ML) techniques in the field of bioelectronics, since bioelectronics and ML are two distinct fields. This article allows bioelectronics researcher to get to know the latest advancement in the ML field. On the other hand, the article provides an insight to the ML researchers about how ML techniques can be useful in bioelectronics applications.

Epilepsy seizure prediction with few-shot learning method

Epileptic seizures prediction and timely alarms allow the patient to take effective and preventive actions. In this paper, a convolutional neural network (CNN) is proposed to diagnose the preictal period. Our goal is for those epileptic patients in whom seizures occur late and it is very challenging to record the preictal signal for them. In the previous works, generalized methods were inevitably used for this group of patients which were not very accurate. Our approach to solve this problem is to provide a few-shot learning method. This method, having the previous knowledge, is trained with only a small number of samples, learns new tasks and reduces the efforts to collect more data. Evaluation results for three patients from the CHB–MIT database, for a 10-min seizure prediction horizon (SPH) and a 20-min seizure occurrence period (SOP), averaged sensitivity of 95.70% and a false prediction rate (FPR) of 0.057/h and for the 5-min prediction horizon and the 25-min seizure occurrence period averaged sensitivity of 98.52% and a false prediction rate of (FPR) of 0.045/h. The proposed few-shot learning method, based on previous knowledge gained from the generalizable method, is regulated with a few new patient samples for the patient. Our results show that the accuracy obtained in this method is higher than the generalizable methods.

Seizure Detection and Prediction by Parallel Memristive Convolutional Neural Networks

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

EEG Peak Detection in Cognitive Conflict Processing Using Summit Navigator and Clustering-based Ranking

Correct detection of peaks in electroencephalogram (EEG) signals is of essence due to the significant correlation of those potentials with cognitive performance and disorders. This paper proposes a novel and non-parametric approach to detect prediction error negativity (PEN) in cognitive conflict processing. The PEN candidates are first located from the input signal via an adaptation of a recent effective method for local maxima extraction, processed in a multi-scale manner. The found candidates are then fused and ranked based on their shape and location-based features. False positives caused by candidates' magnitude are eliminated by rotating the sorted candidate list where the one with the second-best ranking score will be identified as PEN. The EEG data collected from a 3D object selection task have been used to verify the efficacy of the proposed approach. Compared with the state-of-the-art peak detection techniques, the proposed method shows an improvement of at least 2.67% in accuracy and 6.27% in sensitivity while requires only about 4 ms to process an epoch. The accuracy and computational efficiency of the proposed technique in the detection of PEN in cognitive conflict processing would lead to promising applications in performance improvement of brain-computer interfaces (BCIs).

Epileptic electroencephalography classification using Embedded Dynamic Mode Decomposition

OBJECTIVE

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

APPROACH

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

MAIN RESULTS

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

SIGNIFICANCE

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

A novel consistency-based training strategy for seizure prediction

BACKGROUND

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

NEW METHOD

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

RESULTS

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

COMPARISON WITH EXISTING METHODS

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

CONCLUSIONS

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

Semisupervised Seizure Prediction in Scalp EEG Using Consistency Regularization

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

Seizure Prediction in EEG Signals Using STFT and Domain Adaptation

Epileptic seizure prediction is one of the most used therapeutic adjuvant strategies for drug-resistant epilepsy. Conventional approaches commonly collect training and testing samples from the same patient due to inter-individual variability. However, the challenging problem of domain shift between various subjects remains unsolved, resulting in a low conversion rate to the clinic. In this work, a domain adaptation (DA)-based model is proposed to circumvent this issue. The short-time Fourier transform (STFT) is employed to extract the time-frequency features from raw EEG data, and an autoencoder is developed to map these features into high-dimensional space. By minimizing the inter-domain distance in the embedding space, this model learns the domain-invariant information, such that the generalization ability is improved by distribution alignment. Besides, to increase the feasibility of its application, this work mimics the data distribution under the clinical sampling situation and tests the model under this condition, which is the first study that adopts the assessment strategy. Experimental results on both intracranial and scalp EEG databases demonstrate that this method can minimize the domain gap effectively compared with previous approaches.

Power efficient refined seizure prediction algorithm based on an enhanced benchmarking

Deep learning techniques have led to significant advancements in seizure prediction research. However, corresponding used benchmarks are not uniform in published results. Moreover, inappropriate training and evaluation processes used in various work create overfitted models, making prediction performance fluctuate or unreliable. In this study, we analyzed the various data preparation methods, dataset partition methods in related works, and explained the corresponding impacts to the prediction algorithms. Then we applied a robust processing procedure that considers the appropriate sampling parameters and the leave-one-out cross-validation method to avoid possible overfitting and provide prerequisites for ease benchmarking. Moreover, a deep learning architecture takes advantage of a one-dimension convolutional neural network and a bi-directional long short-term memory network is proposed for seizure prediction. The architecture achieves 77.6% accuracy, 82.7% sensitivity, and 72.4% specificity, and it outperforms the indicators of other prior-art works. The proposed model is also hardware friendly; it has 6.274 k parameters and requires only 12.825 M floating-point operations, which is advantageous for memory and power constrained device implementations.

Patient-Independent Seizure Detection Based on Channel-Perturbation Convolutional Neural Network and Bidirectional Long Short-Term Memory

Automatic seizure detection is of great significance for epilepsy diagnosis and alleviating the massive burden caused by manual inspection of long-term EEG. At present, most seizure detection methods are highly patient-dependent and have poor generalization performance. In this study, a novel patient-independent approach is proposed to effectively detect seizure onsets. First, the multi-channel EEG recordings are preprocessed by wavelet decomposition. Then, the Convolutional Neural Network (CNN) with proper depth works as an EEG feature extractor. Next, the obtained features are fed into a Bidirectional Long Short-Term Memory (BiLSTM) network to further capture the temporal variation characteristics. Finally, aiming to reduce the false detection rate (FDR) and improve the sensitivity, the postprocessing, including smoothing and collar, is performed on the outputs of the model. During the training stage, a novel channel perturbation technique is introduced to enhance the model generalization ability. The proposed approach is comprehensively evaluated on the CHB-MIT public scalp EEG database as well as a more challenging SH-SDU scalp EEG database we collected. Segment-based average accuracies of 97.51% and 93.70%, event-based average sensitivities of 86.51% and 89.89%, and average AUC-ROC of 90.82% and 90.75% are yielded on the CHB-MIT database and SH-SDU database, respectively.

Epileptic Seizure Prediction Using Deep Transformer Model

The electroencephalogram (EEG) is the most promising and efficient technique to study epilepsy and record all the electrical activity going in our brain. Automated screening of epilepsy through data-driven algorithms reduces the manual workload of doctors to diagnose epilepsy. New algorithms are biased either towards signal processing or deep learning, which holds subjective advantages and disadvantages. The proposed pipeline is an end-to-end automated seizure prediction framework with a Fourier transform feature extraction and deep learning-based transformer model, a blend of signal processing and deep learning - this imbibes the potential features to automatically identify the attentive regions in EEG signals for effective screening. The proposed pipeline has demonstrated superior performance on the benchmark dataset with average sensitivity and false-positive rate per hour (FPR/h) as 98.46%, 94.83% and 0.12439, 0, respectively. The proposed work shows great results on the benchmark datasets and a big potential for clinics as a support system with medical experts monitoring the patients.

One-Dimensional Convolutional Neural Networks Combined with Channel Selection Strategy for Seizure Prediction Using Long-Term Intracranial EEG

Seizure prediction using intracranial electroencephalogram (iEEG) has attracted an increasing attention during recent years. iEEG signals are commonly recorded in the form of multiple channels. Many previous studies generally used the iEEG signals of all channels to predict seizures, ignoring the consideration of channel selection. In this study, a method of one-dimensional convolutional neural networks (1D-CNN) combined with channel selection strategy was proposed for seizure prediction. First, we used 30-s sliding windows to segment the raw iEEG signals. Then, the 30-s iEEG segments, which were in three channel forms (single channel, channels only from seizure onset or free zone and all channels from seizure onset and free zones), were used as the inputs of 1D-CNN for classification, and the patient-specific model was trained. Finally, the channel form with the best classification was selected for each patient. The proposed method was evaluated on the Freiburg Hospital iEEG dataset. In the situation of seizure occurrence period (SOP) of 30[Formula: see text]min and seizure prediction horizon (SPH) of 5[Formula: see text]min, 98.60[Formula: see text] accuracy, 98.85[Formula: see text] sensitivity and 0.01/h false prediction rate (FPR) were achieved. In the situation of SOP of 60[Formula: see text]min and SPH of 5[Formula: see text]min, 98.32[Formula: see text] accuracy, 98.48[Formula: see text] sensitivity and 0.01/h FPR were attained. Compared with the many existing methods using the same iEEG dataset, our method showed a better performance.

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.

A Hybrid DenseNet-LSTM Model for Epileptic Seizure Prediction

The number of people diagnosed with epilepsy as a common brain disease accounts for about 1% of the world’s total population. Seizure prediction is an important study that can improve the lives of patients with epilepsy, and, in recent years, it has attracted more and more attention. In this paper, we propose a novel hybrid deep learning model that combines a Dense Convolutional Network (DenseNet) and Long Short-Term Memory (LSTM) for epileptic seizure prediction using EEG data. The proposed method first converts the EEG data into the time-frequency domain through Discrete Wavelet Transform (DWT) for use in the input of the model. Then, we train the previously transformed image through a hybrid model combining Densenet and LSTM. To evaluate the performance of the proposed method, experiments are conducted for each preictal length of 5, 10, and 15 min using the CHB-MIT scalp EEG dataset. As a result, we obtained a prediction accuracy of 93.28%, a sensitivity of 92.92%, a specificity of 93.65%, a false positive rate of 0.063 per hour, and an F1-score of 0.923 when the preictal length was 5 min. Finally, as the proposed method is compared to previous studies, it is confirmed that the seizure prediction performance was improved significantly.

A Closed-Loop Neuromodulation Chipset With 2-Level Classification Achieving 1.5-Vpp CM Interference Tolerance, 35-dB Stimulation Artifact Rejection in 0.5ms and 97.8%-Sensitivity Seizure Detection

This work presents an 8-channel closed-loop neuromodulation chipset with 2-level seizure classification. The power-consuming fine classifier is only enabled when the coarse classifier in the frontend chip judges the patient's status as "suspected seizure". This scheme can reduce the overall power consumption extensively since seizure usually occurs with very low possibility. In the capacitive-coupled instrument amplifier (CCIA) of the front-end IC, a feedback based common-mode (CM) cancellation circuit is proposed to suppress large-scale CM interferences and the stimulation artifacts are suppressed by a mixed-signal loop with fast response. An auto-zero based pre- charge path is adopted to boost the input impedance, while the electrode DC offset is canceled by a DC servo loop with very-large and accurate time constant. The 2.32-mm² front-end chip and 3.51-mm² DSP chip implemented in 0.18 μm CMOS are applied in a deep-brain stimulation (DBS) neuromodulator. Measurement results show that the CCIA can suppress 1.5-Vpp CM interference, and achieve an accurate high-pass corner frequency as low as 0.1 Hz and an input impedance greater than 2.2 GΩ. The overall classifier achieves 97.8% sensitivity and consumes only 1.16-μW average power for the CHB-MIT database test. The chipset has been verified by in vivo measurement, showing that the stimulation artifact can be suppressed by 35 dB within 0.5 ms.

Seizure Forecasting Using Long-Term Electroencephalography and Electrocardiogram Data

Electroencephalography (EEG) has been used to forecast seizures with varying success. There is an increasing interest to use electrocardiogram (ECG) to help with seizure forecasting. The neural and cardiovascular systems may exhibit critical slowing, which is measured by an increase in variance and autocorrelation of the system, when change from a normal state to an ictal state. To forecast seizures, the variance and autocorrelation of long-term continuous EEG and ECG data from 16 patients were used for analysis. The average period of recordings was 161.9 h, with an average of 9 electrographic seizures in an individual patient. The relationship between seizure onset times and phases of variance and autocorrelation in EEG and ECG data was investigated. The results of forecasting models using critical slowing features, seizure circadian features, and combined critical slowing and circadian features were compared using the receiver-operating characteristic curve. The results demonstrated that the best forecaster was patient-specific and the average area under the curve (AUC) of the best forecaster across patients was 0.68. In 50% of patients, circadian forecasters had the best performance. Critical slowing forecaster performed best in 19% of patients. Combined forecaster achieved the best performance in 31% of patients. The results of this study may help to advance the field of seizure forecasting and lead to the improved quality of life of people who suffer from epilepsy.

Distributed sensor and actuator networks for closed-loop bioelectronic medicine

Designing implantable bioelectronic systems that continuously monitor physiological functions and simultaneously provide personalized therapeutic solutions for patients remains a persistent challenge across many applications ranging from neural systems to bioelectronic organs. Closed-loop systems typically consist of three functional blocks, namely, sensors, signal processors and actuators. An effective system, that can provide the necessary therapeutics, tailored to individual physiological factors requires a distributed network of sensors and actuators. While significant progress has been made, closed-loop systems still face many challenges before they can truly be considered as long-term solutions for many diseases. In this review, we consider three important criteria where materials play a critical role to enable implantable closed-loop systems: Specificity, Biocompatibility and Connectivity. We look at the progress made in each of these fields with respect to a specific application and outline the challenges in creating bioelectronic technologies for the future.

A Deep Fourier Neural Network for Seizure Prediction Using Convolutional Neural Network and Ratios of Spectral Power

Epileptic seizure prediction is one of the most used therapeutic adjuvant strategies for drug-resistant epilepsy. Conventional methods usually adopt handcrafted features and manual parameter setting. The over-reliance on the expertise of specialists may lead to weak exploitation of features and low popularization of clinical application. This paper proposes a novel parameterless patient-specific method based on Fourier Neural Network (FNN), where the Fourier transform and backpropagation learning are synthesized to make the predictor more efficient and practical. The employment of FNN is the first attempt in the field of seizure prediction due to its automatic extraction of immanent spectra in epileptic signals. Despite the self-adaptive superiority of FNN, we introduce Convolutional Neural Network (CNN) to further improve its search capability in high-dimensional feature spaces. The study also develops a multi-layer module to estimate spectral power ratios of raw recordings, which optimizes the prediction by enhancing feature diversity. Based on these modules, this paper proposes a two-channel deep neural network: Fourier Ratio Convolutional Neural Network (FRCNN). To demonstrate the reliability of the model, we explain the mathematical meaning of hidden-layer neurons in FRCNN theoretically. This approach is evaluated on both intracranial and scalp EEG datasets. It shows that the predictor achieved a sensitivity of 91.2% and a false prediction rate (FPR) of 0.06[Formula: see text]h[Formula: see text] across intracranial subjects and a sensitivity of 85.4% and an FPR of 0.14[Formula: see text]h[Formula: see text] over scalp subjects. The results indicate that FRCNN enables the convenience of epilepsy treatments while preserving a high degree of precision. In the end, a detailed comparison with the previous methods demonstrates that FRCNN has achieved higher performance and generalization ability.

Edge deep learning for neural implants: a case study of seizure detection and prediction

Objective.Implanted devices providing real-time neural activity classification and control are increasingly used to treat neurological disorders, such as epilepsy and Parkinson's disease. Classification performance is critical to identifying brain states appropriate for the therapeutic action (e.g. neural stimulation). However, advanced algorithms that have shown promise in offline studies, in particular deep learning (DL) methods, have not been deployed on resource-restrained neural implants. Here, we designed and optimized three DL models or edge deployment and evaluated their inference performance in a case study of seizure detection.Approach.A deep neural network (DNN), a convolutional neural network (CNN), and a long short-term memory (LSTM) network were designed and trained with TensorFlow to classify ictal, preictal, and interictal phases from the CHB-MIT scalp EEG database. A sliding window based weighted majority voting algorithm was developed to detect seizure events based on each DL model's classification results. After iterative model compression and coefficient quantization, the algorithms were deployed on a general-purpose, off-the-shelf microcontroller for real-time testing. Inference sensitivity, false positive rate (FPR), execution time, memory size, and power consumption were quantified.Main results.For seizure event detection, the sensitivity and FPR for the DNN, CNN, and LSTM models were 87.36%/0.169 h^-1, 96.70%/0.102 h^-1, and 97.61%/0.071 h^-1, respectively. Predicting seizures for early warnings was also feasible. The LSTM model achieved the best overall performance at the expense of the highest power. The DNN model achieved the shortest execution time. The CNN model showed advantages in balanced performance and power with minimum memory requirement. The implemented model compression and quantization achieved a significant saving of power and memory with an accuracy degradation of less than 0.5%.Significance.Inference with embedded DL models achieved performance comparable to many prior implementations that had no time or computational resource limitations. Generic microcontrollers can provide the required memory and computational resources, while model designs can be migrated to application-specific integrated circuits for further optimization and power saving. The results suggest that edge DL inference is a feasible option for future neural implants to improve classification performance and therapeutic outcomes.

Machine Learning for Predicting Epileptic Seizures Using EEG Signals: A Review

With the advancement in artificial intelligence (AI) and machine learning (ML) techniques, researchers are striving towards employing these techniques for advancing clinical practice. One of the key objectives in healthcare is the early detection and prediction of disease to timely provide preventive interventions. This is especially the case for epilepsy, which is characterized by recurrent and unpredictable seizures. Patients can be relieved from the adverse consequences of epileptic seizures if it could somehow be predicted in advance. Despite decades of research, seizure prediction remains an unsolved problem. This is likely to remain at least partly because of the inadequate amount of data to resolve the problem. There have been exciting new developments in ML-based algorithms that have the potential to deliver a paradigm shift in the early and accurate prediction of epileptic seizures. Here we provide a comprehensive review of state-of-the-art ML techniques in early prediction of seizures using EEG signals. We will identify the gaps, challenges, and pitfalls in the current research and recommend future directions.

Comparison of Smoothing Filters' Influence on Quality of Data Recorded with the Emotiv EPOC Flex Brain-Computer Interface Headset during Audio Stimulation

Off-the-shelf, consumer-grade EEG equipment is nowadays becoming the first-choice equipment for many scientists when it comes to recording brain waves for research purposes. On one hand, this is perfectly understandable due to its availability and relatively low cost (especially in comparison to some clinical-level EEG devices), but, on the other hand, quality of the recorded signals is gradually increasing and reaching levels that were offered just a few years ago by much more expensive devices used in medicine for diagnostic purposes. In many cases, a well-designed filter and/or a well-thought signal acquisition method improve the signal quality to the level that it becomes good enough to become subject of further analysis allowing to formulate some valid scientific theories and draw far-fetched conclusions related to human brain operation. In this paper, we propose a smoothing filter based upon the Savitzky-Golay filter for the purpose of EEG signal filtering. Additionally, we provide a summary and comparison of the applied filter to some other approaches to EEG data filtering. All the analyzed signals were acquired from subjects performing visually involving high-concentration tasks with audio stimuli using Emotiv EPOC Flex equipment.

Epileptic seizure recognition using EEG wavelet decomposition based on nonlinear and statistical features with support vector machine classification

Epileptic seizure (ES) is a neurological brain dysfunction. ES can be detected using the electroencephalogram (EEG) signal. However, visual inspection of ES using long-time EEG recordings is a difficult, time-consuming and a costly procedure. Thus, automatic epilepsy recognition is of primary importance. In this paper, a new method is proposed for automatic ES recognition using short-time EEG recordings. The method is based on first decomposing the EEG signals on sub-signals using discrete wavelet transform. Then, from the obtained sub-signals, different non-linear parameters such as approximate entropy (ApEn), largest Lyapunov exponents (LLE) and statistical parameters are determined. These parameters along with phase entropies, calculated through higher order spectrum analysis, are used as an input vector of a multi-class support vector machine (MSVM) for ES recognition. The proposed method is evaluated using the standard EEG database developed by the Department of Epileptology, University of Bonn, Germany. The evaluation is carried out through a large number of classification experiments. Different statistical metrics namely Sensitivity (Se), Specificity (Sp) and classification accuracy (Ac) are calculated and compared to those obtained in the scientific research literature. The obtained results show that the proposed method achieves high accuracies, which are as good as the best existing state-of-the-art methods studied using the same EEG database.

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

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

Epileptic seizure prediction based on EEG spikes detection of ictal-preictal states

Epileptic seizures are known for their unpredictable nature. However, recent research provides that the transition to seizure event is not random but the result of evidence accumulations. Therefore, a reliable method capable to detect these indications can predict seizures and improve the life quality of epileptic patients. Seizures periods are generally characterized by epileptiform discharges with different changes including spike rate variation according to the shapes, spikes, and the amplitude. In this study, spike rate is used as the indicator to anticipate seizures in electroencephalogram (EEG) signal. Spikes detection step is used in EEG signal during interictal, preictal, and ictal periods followed by a mean filter to smooth the spike number. The maximum spike rate in interictal periods is used as an indicator to predict seizures. When the spike number in the preictal period exceeds the threshold, an alarm is triggered. Using the CHB-MIT database, the proposed approach has ensured 92% accuracy in seizure prediction for all patients.

Automatic Seizure Detection Based on S-Transform and Deep Convolutional Neural Network

Automatic seizure detection is significant for the diagnosis of epilepsy and reducing the massive workload of reviewing continuous EEGs. In this work, a novel approach, combining Stockwell transform (S-transform) with deep Convolutional Neural Networks (CNN), is proposed to detect seizure onsets in long-term intracranial EEG recordings. Primarily, raw EEG data is filtered with wavelet decomposition. Then, S-transform is used to obtain a proper time-frequency representation of each EEG segment. After that, a 15-layer deep CNN using dropout and batch normalization serves as a robust feature extractor and classifier. Finally, smoothing and collar technique are applied to the outputs of CNN to improve the detection accuracy and reduce the false detection rate (FDR). The segment-based and event-based evaluation assessments and receiver operating characteristic (ROC) curves are employed for the performance evaluation on a public EEG database containing 21 patients. A segment-based sensitivity of 97.01% and a specificity of 98.12% are yielded. For the event-based assessment, this method achieves a sensitivity of 95.45% with an FDR of 0.36/h.

Epileptic Seizure Detection System Based on Multi-Domain Feature and Spike Feature of EEG

Epilepsy is caused by sudden abnormal discharges of neurons in the brain. This paper constructs an automatic seizure detection system, which combines the predicting result of multi-domain feature with the predicting result of spike rate feature to detect the occurrence of epileptic seizures. After segmenting EEG data into 5[Formula: see text]s with 80% overlap epochs, the paper extracts time domain features, frequency domain features and hurst exponents (HE) from each epoch and these features are reduced by linear discriminant analysis (LDA) to be input parameters of the random forest (RF) classifier, which provides classification of the EEG epochs concerning the existence of seizures. In parallel, the paper extracts spikes from EEG data with morphological filter and calculates the spike rate to determine whether there is seizure. Then the results obtained by these two methods are merged as the final detection result. The paper shows that the accuracy (AC), sensitivity (SE), specificity (SP) and the false positive ratio based on event (FPRE) obtained by hybrid method are 98.94%, 76.60%, 98.99% and 2.43 times/h, respectively. Finally, the paper applies the seizure detection method to do seizure warning and recording to help the family member to take care of the patients and the doctor to adjust the antiepileptic drugs (AEDs).

Real-time epileptic seizure prediction based on online monitoring of pre-ictal features

Reliable prediction of epileptic seizures is of prime importance as it can drastically change the quality of life for patients. This study aims to propose a real-time low computational approach for the prediction of epileptic seizures and to present an efficient hardware implementation of this approach for portable prediction systems. Three levels of feature extraction are performed to characterize the pre-ictal activities of the EEG signal. In the first-level, the line length algorithm is applied to the pre-ictal region. The features obtained in the first-level are mathematically integrated to extract the second-level features and then the line lengths of the second-level features are calculated to obtain our third-level feature. The third-level information is compared with predefined threshold levels to make a decision on whether the extracted characteristics are relevant to a seizure occurrence or not. The validity of this algorithm was tested by EEG recordings in the CHB-MIT database (97 seizures, 834.224 h) for 19 epileptic patients. The results showed that the average sensitivity was 90.62%, the specificity was 88.34%, the accuracy was 88.76% with the average false prediction rate as low as 0.0046 h^-1, and the average prediction time was 23.3 min. The low computational complexity is the superiority of the proposed approach, which provides a technologically simple but accurate way of predicting epileptic seizures and enables hardware implantable devices. Graphical abstract Proposed seizure prediction algorithm and its features.

Seizure prediction - ready for a new era

Epilepsy is a common disorder characterized by recurrent seizures. An overwhelming majority of people with epilepsy regard the unpredictability of seizures as a major issue. More than 30 years of international effort have been devoted to the prediction of seizures, aiming to remove the burden of unpredictability and to couple novel, time-specific treatment to seizure prediction technology. A highly influential review published in 2007 concluded that insufficient evidence indicated that seizures could be predicted. Since then, several advances have been made, including successful prospective seizure prediction using intracranial EEG in a small number of people in a trial of a real-time seizure prediction device. In this Review, we examine advances in the field, including EEG databases, seizure prediction competitions, the prospective trial mentioned and advances in our understanding of the mechanisms of seizures. We argue that these advances, together with statistical evaluations, set the stage for a resurgence in efforts towards the development of seizure prediction methodologies. We propose new avenues of investigation involving a synergy between mechanisms, models, data, devices and algorithms and refine the existing guidelines for the development of seizure prediction technology to instigate development of a solution that removes the burden of the unpredictability of seizures.

Epilepsy Seizure Prediction on EEG Using Common Spatial Pattern and Convolutional Neural Network

Epilepsy seizure prediction paves the way of timely warning for patients to take more active and effective intervention measures. Compared to seizure detection that only identifies the inter-ictal state and the ictal state, far fewer researches have been conducted on seizure prediction because the high similarity makes it challenging to distinguish between the pre-ictal state and the inter-ictal state. In this paper, a novel solution on seizure prediction is proposed using common spatial pattern (CSP) and convolutional neural network (CNN). Firstly, artificial pre-ictal EEG signals based on the original ones are generated by combining the segmented pre-ictal signals to solve the trial imbalance problem between the two states. Secondly, a feature extractor employing wavelet packet decomposition and CSP is designed to extract the distinguishing features in both the time domain and the frequency domain. It can improve overall accuracy while reducing the training time. Finally, a shallow CNN is applied to discriminate between the pre-ictal state and the inter-ictal state. Our proposed solution is evaluated on 23 patients' data from Boston Children's Hospital-MIT scalp EEG dataset by employing a leave-one-out cross-validation, and it achieves a sensitivity of 92.2% and false prediction rate of 0.12/h. Experimental result demonstrates that the proposed approach outperforms most state-of-the-art methods.

Automated epileptic seizure detection based on break of excitation/inhibition balance

Physiological models are attractive for seizure detection, as their parameters are related to physiological meanings. We propose an algorithm to early detect epileptic seizures based on automatic estimation of average synaptic gains (excitatory Ae, slow and fast inhibitory B and G) by combining clinical data with a neural mass model. Three indices (Ae/B, Ae/G and Ae/(B + G)), all related to excitation/inhibition balance, were calculated and used as cues to detect seizures. A simple thresholding method was employed. We evaluated the algorithm against the manual scoring of a human expert on intracranial EEG samples from 23 patients suffering from different types of epilepsy. Best performance was achieved using Ae/(B + G) as a cue, i.e. excitation/(slow + fast) inhibition, on temporal lobe epilepsy (TLE) patients. A leave-one-out cross-validation showed that the algorithm achieved 92.98% sensitivity for TLE patients. The median false positive rate was 0.16 per hour, and median detection delay was 14.5 s. Of interest, the threshold values determined by a leave-one-out cross-validation did nearly not vary among TLE patients, suggesting a general excitation/inhibition balance baseline in TLE patients. The same approach could be used with other types of epilepsy by adapting the neural mass model to these types.

Automated seizure prediction

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

Predicting state transitions in brain dynamics through spectral difference of phase-space graphs

Networks are naturally occurring phenomena that are studied across many disciplines. The topological features of a network can provide insight into the dynamics of a system as it evolves, and can be used to predict changes in state. The brain is a complex network whose temporal and spatial behavior can be measured using electroencephalography (EEG). This data can be reconstructed to form a family of graphs that represent the state of the brain over time, and the evolution of these graphs can be used to predict changes in brain states, such as the transition from preictal to ictal in patients with epilepsy. This research proposes objective indications of seizure onset observed from minimally invasive scalp EEG. The approach considers the brain as a complex nonlinear dynamical system whose state can be derived through time-delay embedding of the EEG data and characterized to determine change in brain dynamics related to the preictal state. This method targets phase-space graph spectra as biomarkers for seizure prediction, correlates historical degrees of change in spectra, and makes accurate prediction of seizure onset. A significant trend of normalized dissimilarity over time indicates a departure from the norm, and thus a change in state. Our methods show high sensitivity (90-100%) and specificity (90%) on 241 h of scalp EEG training data, and sensitivity and specificity of 70%-90% on test data. Moreover, the algorithm was capable of processing 12.7 min of data per second on an Intel Core i3 CPU in Matlab, showing that real-time analysis is viable.