Patient-specific Seizure Prediction Research Articles

On the performance of seizure prediction machine learning methods across different databases: the sample and alarm-based perspectives.

Epilepsy affects 1% of the global population, with approximately one-third of patients resistant to anti-seizure medications (ASMs), posing risks of physical injuries and psychological issues. Seizure prediction algorithms aim to enhance the quality of life for these individuals by providing timely alerts. This study presents a patient-specific seizure prediction algorithm applied to diverse databases (EPILEPSIAE, CHB-MIT, AES, and Epilepsy Ecosystem). The proposed algorithm undergoes a standardized framework, including data preprocessing, feature extraction, training, testing, and postprocessing. Various databases necessitate adaptations in the algorithm, considering differences in data availability and characteristics. The algorithm exhibited variable performance across databases, taking into account sensitivity, FPR/h, specificity, and AUC score. This study distinguishes between sample-based approaches, which often yield better results by disregarding the temporal aspect of seizures, and alarm-based approaches, which aim to simulate real-life conditions but produce less favorable outcomes. Statistical assessment reveals challenges in surpassing chance levels, emphasizing the rarity of seizure events. Comparative analyses with existing studies highlight the complexity of standardized assessments, given diverse methodologies and dataset variations. Rigorous methodologies aiming to simulate real-life conditions produce less favorable outcomes, emphasizing the importance of realistic assumptions and comprehensive, long-term, and systematically structured datasets for future research.

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

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

Addressing data limitations in seizure prediction through transfer learning

According to the literature, seizure prediction models should be developed following a patient-specific approach. However, seizures are usually very rare events, meaning the number of events that may be used to optimise seizure prediction approaches is limited. To overcome such constraint, we analysed the possibility of using data from patients from an external database to improve patient-specific seizure prediction models. We present seizure prediction models trained using a transfer learning procedure. We trained a deep convolutional autoencoder using electroencephalogram data from 41 patients collected from the EPILEPSIAE database. Then, a bidirectional long short-term memory and a classifier layers were added on the top of the encoder part and were optimised for 24 patients from the Universitätsklinikum Freiburg individually. The encoder was used as a feature extraction module. Therefore, its weights were not changed during the patient-specific training. Experimental results showed that seizure prediction models optimised using pretrained weights present about four times fewer false alarms while maintaining the same ability to predict seizures and achieved more 13% validated patients. Therefore, results evidenced that the optimisation using transfer learning was more stable and faster, saving computational resources. In summary, adopting transfer learning for seizure prediction models represents a significant advancement. It addresses the data limitation seen in the seizure prediction field and offers more efficient and stable training, conserving computational resources. Additionally, despite the compact size, transfer learning allows to easily share data knowledge due to fewer ethical restrictions and lower storage requirements. The convolutional autoencoder developed in this study will be shared with the scientific community, promoting further research.

Concept-drifts adaptation for machine learning EEG epilepsy seizure prediction.

Seizure prediction remains a challenge, with approximately 30% of patients unresponsive to conventional treatments. Addressing this issue is crucial for improving patients' quality of life, as timely intervention can mitigate the impact of seizures. In this research field, it is critical to identify the preictal interval, the transition from regular brain activity to a seizure. While previous studies have explored various Electroencephalogram (EEG) based methodologies for prediction, few have been clinically applicable. Recent studies have underlined the dynamic nature of EEG data, characterised by data changes with time, known as concept drifts, highlighting the need for automated methods to detect and adapt to these changes. In this study, we investigate the effectiveness of automatic concept drift adaptation methods in seizure prediction. Three patient-specific seizure prediction approaches with a 10-minute prediction horizon are compared: a seizure prediction algorithm incorporating a window adjustment method by optimising performance with Support Vector Machines (Backwards-Landmark Window), a seizure prediction algorithm incorporating a data-batch (seizures) selection method using a logistic regression (Seizure-batch Regression), and a seizure prediction algorithm with a dynamic integration of classifiers (Dynamic Weighted Ensemble). These methods incorporate a retraining process after each seizure and use a combination of univariate linear features and SVM classifiers. The Firing Power was used as a post-processing technique to generate alarms before seizures. These methodologies were compared with a control approach based on the typical machine learning pipeline, considering a group of 37 patients with Temporal Lobe Epilepsy from the EPILEPSIAE database. The best-performing approach (Backwards-Landmark Window) achieved results of 0.75 ± 0.33 for sensitivity and 1.03 ± 1.00 for false positive rate per hour. This new strategy performed above chance for 89% of patients with the surrogate predictor, whereas the control approach only validated 46%.

Seizure prediction based on improved vision transformer model for EEG channel optimization

Epileptic seizures are unpredictable events caused by abnormal discharges of a patient’s brain cells. Extensive research has been conducted to develop seizure prediction algorithms based on long-term continuous electroencephalogram (EEG) signals. This paper describes a patient-specific seizure prediction method that can serve as a basis for the design of lightweight, wearable and effective seizure-prediction devices. We aim to achieve two objectives using this method. The first aim is to extract robust feature representations from multichannel EEG signals, and the second aim is to reduce the number of channels used for prediction by selecting an optimal set of channels from multichannel EEG signals while ensuring good prediction performance. We design a seizure-prediction algorithm based on a vision transformer (ViT) model. The algorithm selects channels that play a key role in seizure prediction from 22 channels of EEG signals. First, we perform a time-frequency analysis of processed time-series signals to obtain EEG spectrograms. We then segment the spectrograms of multiple channels into many non-overlapping patches of the same size, which are input into the channel selection layer of the proposed model, named Sel-JPM-ViT, enabling it to select channels. Application of the Sel-JPM-ViT model to the Boston Children’s Hospital–Massachusetts Institute of Technology scalp EEG dataset yields results using only three to six channels of EEG signals that are slightly better that the results obtained using 22 channels of EEG signals. Overall, the Sel-JPM-ViT model exhibits an average classification accuracy of 93.65%, an average sensitivity of 94.70% and an average specificity of 92.78%.

EEG epilepsy seizure prediction: the post-processing stage as a chronology

Almost one-third of epileptic patients fail to achieve seizure control through anti-epileptic drug administration. In the scarcity of completely controlling a patient’s epilepsy, seizure prediction plays a significant role in clinical management and providing new therapeutic options such as warning or intervention devices. Seizure prediction algorithms aim to identify the preictal period that Electroencephalogram (EEG) signals can capture. However, this period is associated with substantial heterogeneity, varying among patients or even between seizures from the same patient. The present work proposes a patient-specific seizure prediction algorithm using post-processing techniques to explore the existence of a set of chronological events of brain activity that precedes epileptic seizures. The study was conducted with 37 patients with Temporal Lobe Epilepsy (TLE) from the EPILEPSIAE database. The designed methodology combines univariate linear features with a classifier based on Support Vector Machines (SVM) and two post-processing techniques to handle pre-seizure temporality in an easily explainable way, employing knowledge from network theory. In the Chronological Firing Power approach, we considered the preictal as a sequence of three brain activity events separated in time. In the Cumulative Firing Power approach, we assumed the preictal period as a sequence of three overlapping events. These methodologies were compared with a control approach based on the typical machine learning pipeline. We considered a Seizure Prediction horizon (SPH) of 5 mins and analyzed several values for the Seizure Occurrence Period (SOP) duration, between 10 and 55 mins. Our results showed that the Cumulative Firing Power approach may improve the seizure prediction performance. This new strategy performed above chance for 62% of patients, whereas the control approach only validated 49% of its models.

Hardware implementation of deep neural network for seizure prediction

Epilepsy is a neurological disorder characterized by seizures, which are caused by a sudden, uncontrollable electrical disturbance in the brain. Recently, machine learning and deep learning techniques have been used in seizure prediction algorithms, which greatly aids epilepsy patients. Moreover, the usage of hardware that is available for both medical practitioners and epilepsy patients would help enhance the quality of life for epilepsy patients. For this purpose, a Field-programmable Gate Array (FPGA) is used in this work to implement a hardware model of a neural network that is able to predict seizures. In this research, the main aim is to implement an FPGA-based general and patient-specific seizure prediction algorithm that detects seizures for epilepsy patients using Multilayer Perceptron (MLP) neural network models. Moreover, the work will also tackle ways to optimize the FPGA resources and the computational time of the seizure prediction models. The data available for training and testing are raw electroencephalogram (EEG) signal samples provided by the Melbourne-NeuroVista seizure trial and the Melbourne-University AES-MathWorks-NIH Seizure Prediction Challenge. The results show that the general model had an Area under curve (AUC) score of 0.72 while the patient-specific model had an average AUC score of 0.8.

Cross-Domain Transfer of EEG to EEG or ECG Learning for CNN Classification Models.

Electroencephalography (EEG) is often used to evaluate several types of neurological brain disorders because of its noninvasive and high temporal resolution. In contrast to electrocardiography (ECG), EEG can be uncomfortable and inconvenient for patients. Moreover, deep-learning techniques require a large dataset and a long time for training from scratch. Therefore, in this study, EEG-EEG or EEG-ECG transfer learning strategies were applied to explore their effectiveness for the training of simple cross-domain convolutional neural networks (CNNs) used in seizure prediction and sleep staging systems, respectively. The seizure model detected interictal and preictal periods, whereas the sleep staging model classified signals into five stages. The patient-specific seizure prediction model with six frozen layers achieved 100% accuracy for seven out of nine patients and required only 40 s of training time for personalization. Moreover, the cross-signal transfer learning EEG-ECG model for sleep staging achieved an accuracy approximately 2.5% higher than that of the ECG model; additionally, the training time was reduced by >50%. In summary, transfer learning from an EEG model to produce personalized models for a more convenient signal can both reduce the training time and increase the accuracy; moreover, challenges such as data insufficiency, variability, and inefficiency can be effectively overcome.

Construction of an epileptic seizure prediction model using a semi-supervised method of generative adversarial and long short term memory network combined with Stockwell transform

To propose a semi-supervised epileptic seizure prediction model (ST-WGAN-GP-Bi-LSTM) to enhance the prediction performance by improving time-frequency analysis of electroencephalogram (EEG) signals, enhancing the stability of the unsupervised feature learning model and improving the design of back-end classifier. Stockwell transform (ST) of the epileptic EEG signals was performed to locate the time-frequency information by adaptive adjustment of the resolution and retaining the absolute phase to obtain the time-frequency inputs. When there was no overlap between the generated data distribution and the real EEG data distribution, to avoid failure of feature learning due to a constant JS divergence, Wasserstein GAN was used as a feature learning model, and the cost function based on EM distance and gradient penalty strategy was adopted to constrain the unsupervised training process to allow the generation of a high-order feature extractor. A temporal prediction model was finally constructed based on a bi-directional long short term memory network (Bi-LSTM), and the classification performance was improved by obtaining the temporal correlation between high-order time-frequency features. The CHB-MIT scalp EEG dataset was used to validate the proposed patient-specific seizure prediction method. The AUC, sensitivity, and specificity of the proposed method reached 90.40%, 83.62%, and 86.69%, respectively. Compared with the existing semi-supervised methods, the propose method improved the original performance by 17.77%, 15.41%, and 53.66%. The performance of this method was comparable to that of a supervised prediction model based on CNN. The utilization of ST, WGAN-GP, and Bi-LSTM effectively improves the prediction performance of the semi-supervised deep learning model, which can be used for optimization of unsupervised feature extraction in epileptic seizure prediction.

Dynamic Multi-Graph Convolution based Channel-Weighted Transformer Feature Fusion Network for Epileptic Seizure Prediction.

Electroencephalogram (EEG) based seizure prediction plays an important role in the closed-loop neuromodulation system. However, most existing seizure prediction methods based on graph convolution network only focused on constructing the static graph, ignoring multi-domain dynamic changes in deep graph structure. Moreover, the existing feature fusion strategies generally concatenated coarse-grained epileptic EEG features directly, leading to the suboptimal seizure prediction performance. To address these issues, we propose a novel multi-branch dynamic multi-graph convolution based channel-weighted transformer feature fusion network (MB-dMGC-CWTFFNet) for the patient-specific seizure prediction with the superior performance. Specifically, a multi-branch (MB) feature extractor is first applied to capture the temporal, spatial and spectral representations fromthe epileptic EEG jointly. Then, we design a point-wise dynamic multi-graph convolution network (dMGCN) to dynamically learn deep graph structures, which can effectively extract high-level features from the multi-domain graph. Finally, by integrating the local and global channel-weighted strategies with the multi-head self-attention mechanism, a channel-weighted transformer feature fusion network (CWTFFNet) is adopted to efficiently fuse the multi-domain graph features. The proposed MB-dMGC-CWTFFNet is evaluated on the public CHB-MIT EEG and a private intracranial sEEG datasets, and the experimental results demonstrate that our proposed method achieves outstanding prediction performance compared with the state-of-the-art methods, indicating an effective tool for patient-specific seizure warning. Our code will be available at: https://github.com/Rockingsnow/MB-dMGC-CWTFFNet.

EEG-Based Seizure Prediction via Model Uncertainty Learning.

Deep neural networks (DNNs) have the powerful ability to automatically extract efficient features, which makes them prominent in electroencephalogram (EEG) based seizure prediction tasks. However, current research in this field cannot take the model uncertainty into account, causing the prediction less credible. To this end, we introduce a novel end-to-end patient-specific seizure prediction framework via model uncertainty learning. Specifically, we propose a reparameterized EEG-based lightweight CNN architecture and a modified Monte Carlo dropout (RepNet-MMCD) strategy to improve the reliability of the DNNs-based model. In RepNet, we obtain multi-scale feature representations by applying depthwise separable convolutions of different kernels. After training, depthwise convolutions with different scales are equivalently converted into a single convolution layer, which can greatly reduce computational budgets without losing model performance. In addition, we propose a modified Monte Carlo (MMCD) strategy, leveraging the samples-based temporal information in EEG signals to simulate the Monte Carlo dropout sampling. Sensitivity, false-positive rate (FPR), and area under curve (AUC) of the proposed RepNet-MMCD achieve 93.1%, 0.033/h, 0.950 and 81.6%, 0.056/h, 0.903 on two public datasets, respectively. We further extend the MMCD strategy to the other baseline methods, which can improve the performance of seizure prediction by a clear margin.

Patient-specific method for predicting epileptic seizures based on DRSN-GRU

Epilepsy is one of the most common neurological disorders worldwide and can cause the brain to stop working properly or even endanger the life of the patient. Epilepsy prediction is a prerequisite for seizure control, allowing preventive measures to mitigate damage or control seizures. It has been found that abnormal brain activity begins some time before a seizure, which is known as the pre-ictal state. In this study, we reconsidered the temporal scope of the pre-ictal period and divided it into multiple temporal windows. A patient-specific seizure prediction method based on deep residual shrinkage network (DRSN) and gated recurrent unit (GRU) was then proposed. The temporal dependency of the signal of different time windows in the pre-ictal period is modeled by introducing GRU into a DRSN. In addition, automatic feature extraction is achieved by applying soft threshold denoising and attention mechanism inside the neural network. The proposed method was tested on four patients reasonably selected from the CHB-MIT scalp EEG dataset, which achieved a sensitivity of 90.54%, an AUC value of 0.88, and a false prediction rate of 0.11/h. The results obtained by our method are compared with the recent epilepsy prediction methods. Compared with the being among the best method, our method still has a little gap, but it also shows a new idea and some advantages.

Patient-Specific Seizure Prediction via Adder Network and Supervised Contrastive Learning.

Deep learning (DL) methods have been widely used in the field of seizure prediction from electroencephalogram (EEG) in recent years. However, DL methods usually have numerous multiplication operations resulting in high computational complexity. In addtion, most of the current approaches in this field focus on designing models with special architectures to learn representations, ignoring the use of intrinsic patterns in the data. In this study, we propose a simple and effective end-to-end adder network and supervised contrastive learning (AddNet-SCL). The method uses addition instead of the massive multiplication in the convolution process to reduce the computational cost. Besides, contrastive learning is employed to effectively use label information, points of the same class are clustered together in the projection space, and points of different class are pushed apart at the same time. Moreover, the proposed model is trained by combining the supervised contrastive loss from the projection layer and the cross-entropy loss from the classification layer. Since the adder networks uses the l1 -norm distance as the similarity measure between the input feature and the filters, the gradient function of the network changes, an adaptive learning rate strategy is employed to ensure the convergence of AddNet-CL. Experimental results show that the proposed method achieves 94.9% sensitivity, an area under curve (AUC) of 94.2%, and a false positive rate of (FPR) 0.077/h on 19 patients in the CHB-MIT database and 89.1% sensitivity, an AUC of 83.1%, and an FPR of 0.120/h in the Kaggle database. Competitive results show that this method has broad prospects in clinical practice.

Pediatric Seizure Prediction in Scalp EEG Using a Multi-Scale Neural Network With Dilated Convolutions.

Objective: Epileptic seizure prediction based on scalp electroencephalogram (EEG) is of great significance for improving the quality of life of patients with epilepsy. In recent years, a number of studies based on deep learning methods have been proposed to address this issue and achieve excellent performance. However, most studies on epileptic seizure prediction by EEG fail to take full advantage of temporal-spatial multi-scale features of EEG signals, while EEG signals carry information in multiple temporal and spatial scales. To this end, in this study, we proposed an end-to-end framework by using a temporal-spatial multi-scale convolutional neural network with dilated convolutions for patient-specific seizure prediction. Methods: Specifically, the model divides the EEG processing pipeline into two stages: the temporal multi-scale stage and the spatial multi-scale stage. In each stage, we firstly extract the multi-scale features along the corresponding dimension. A dilated convolution block is then conducted on these features to expand our model’s receptive fields further and systematically aggregate global information. Furthermore, we adopt a feature-weighted fusion strategy based on an attention mechanism to achieve better feature fusion and eliminate redundancy in the dilated convolution block. Results: The proposed model obtains an average sensitivity of 93.3%, an average false prediction rate of 0.007 per hour, and an average proportion of time-in-warning of 6.3% testing in 16 patients from the CHB-MIT dataset with the leave-one-out method. Conclusion: Our model achieves superior performance in comparison to state-of-the-art methods, providing a promising solution for EEG-based seizure prediction.

Seizure Prediction Using Directed Transfer Function and Convolution Neural Network on Intracranial EEG.

Automatic seizure prediction promotes the development of closed-loop treatment system on intractable epilepsy. In this study, by considering the specific information exchange between EEG channels from the perspective of whole brain activities, the convolution neural network (CNN) and the directed transfer function (DTF) were merged to present a novel method for patient-specific seizure prediction. Firstly, the intracranial electroencephalogram (iEEG) signals were segmented and the information flow features of iEEG signals were calculated by using the DTF algorithm. Then, these features were reconstructed as the channel-frequency maps according to channel pairs and the frequency of information flow. Finally, these maps were fed into the CNN model and the outputs were post-processed by the moving average approach to predict the epileptic seizures. By the evaluation of cross-validation method, the proposed algorithm achieved the averaged sensitivity of 90.8%, the averaged false prediction rate of 0.08 per hour. Compared to the random predictor and other existing algorithms tested on the Freiburg EEG dataset, our proposed method achieved better performance for seizure prediction in all patients. These results demonstrated that the proposed algorithm could provide an robust seizure prediction solution by using deep learning to capture the brain network changes of iEEG signals from epileptic patients.

Roughness-Length-Based Characteristic Analysis of Intracranial EEG and Epileptic Seizure Prediction.

To identify precursors of epileptic seizures, an EEG characteristic analysis is carried out based on a roughness-length method, where fractal dimensions and intercept values are extracted to measure the structure complexity and the amplitude roughness of EEG signals in different phases. Using the significant changes of the fractal dimension and intercept in the preictal phase with respect to those in the interictal phase, a patient-specific seizure prediction algorithm is then proposed by combining with a gradient boosting classifier. The probabilistic outputs of the trained gradient boosting classifier are further processed by threshold comparison and rule-based judgment to distinguish preictal EEG from interictal EEG and to generate seizure alerts. The prediction algorithm was evaluated on 20 patients' intracranial EEG recordings from the Freiburg EEG database, which contains the preictal periods of 65 seizures and 499[Formula: see text]h interictal EEG. Setting the seizure prediction horizon as 2[Formula: see text]min, averaged sensitivity values of 90.42% and 91.67% with averaged false prediction rates of 0.12/h and 0.10/h were achieved for seizure occurrence periods of 30 and 50[Formula: see text]min, respectively. These results demonstrate the ability of fractal dimension and intercept metrics in predicting the occurrence of seizures.

EEG Synchronization Analysis for Seizure Prediction: A Study on Data of Noninvasive Recordings

Objective: Epilepsy is a neurological disorder arising from anomalies of the electrical activity in the brain, affecting ~65 million individuals worldwide. Prediction methods, typically based on machine learning methods, require a large amount of data for training, in order to correctly classify seizures with small false alarm rates. Methods: In this work, we present a new database containing EEG scalp signals of 14 epileptic patients acquired at the Unit of Neurology and Neurophysiology of the University of Siena, Italy. Furthermore, a patient-specific seizure prediction method, based on the detection of synchronization patterns in the EEG, is proposed and tested on the data of the database. The use of noninvasive EEG data aims to explore the possibility of developing a noninvasive monitoring/control device for the prediction of seizures. The prediction method employs synchronization measures computed over all channel pairs and a computationally inexpensive threshold-based classification approach. Results and conclusions: The experimental analysis, performed by inspection and by the proposed threshold-based classifier on all the patients of the database, shows that the features extracted by the synchronization measures are able to detect preictal and ictal states and allow the prediction of the seizures few minutes before the seizure onsets.

Seizure Prediction Using Multi-View Features and Improved Convolutional Gated Recurrent Network

Epilepsy is one of the most common neurological diseases worldwide. Early prediction of seizure onsets is of great significance for the safety of intractable epilepsy patients. This work aims to develop a reliable and accurate method for patient-specific seizure prediction based on scalp electroencephalograms (EEGs). Local fractal spectrum, relative band energy, and synchronization modularity features are used to reveal the characteristics of multi-channel EEG in perspectives of time domain, frequency domain, and functional connectivity, respectively. A novel framework, named multi-view convolutional gated recurrent network (Mv-CGRN), is proposed to comprehensively analyze the spatio-temporal sequences of multi-view features and capture the potential variations preceding the impending seizure. Moreover, an attention mechanism is embedded in Mv-CGRN to determine the optimal feature combinations for each patient by adaptively tuning the weight parameters. The proposed system achieves an average sensitivity of 94.50% and an average false positive rate (FPR) of 0.118/h on CHB-MIT scalp EEG dataset, using the leave-one-out cross validation (LOOCV). Our work shows a promising performance compared with the state-of-the-art works in the same filed.

Seizure Prediction in Scalp EEG Using 3D Convolutional Neural Networks With an Image-Based Approach.

Epileptic seizures occur as a result of a process that develops over time and space in epileptic networks. In this study, we aim at developing a generalizable method for patient-specific seizure prediction by evaluating the spatio-temporal correlation in the features obtained from multichannel EEG signals. Spectral band power, statistical moment and Hjorth parameters are used to reveal the frequency and time domain features of the EEG signals. The features are given as input to a convolutional neural network (CNN) by transforming into a sequence of multi-color images according to the topology of the EEG channels. The multi-frame 3D CNN model is proposed to evaluate the temporal and spatial correlation in training data collectively. The proposed 3D CNN model achieves a sensitivity of 85.7%, a false prediction rate of 0.096/h, and a proportion of time-in-warning of 10.5%, in the tests performed with 16 patients from the CHB-MIT scalp EEG dataset. The results show that the superiority of the proposed method to a Poisson based random predictor was statistically significant for 93.7% of the patients, at significance level of 0.05. Our experiments with various timing constraints show that epileptic stage lengths are an important factor affecting seizure performance. We present a subject-specific seizure prediction method that is robust for unbalanced data and can be generalized to any scalp EEG dataset without the need for subject-specific engineering.

Efficient Epileptic Seizure Prediction Based on Deep Learning

Epilepsy is one of the world's most common neurological diseases. Early prediction of the incoming seizures has a great influence on epileptic patients' life. In this paper, a novel patient-specific seizure prediction technique based on deep learning and applied to long-term scalp electroencephalogram (EEG) recordings is proposed. The goal is to accurately detect the preictal brain state and differentiate it from the prevailing interictal state as early as possible and make it suitable for real time. The features extraction and classification processes are combined into a single automated system. Raw EEG signal without any preprocessing is considered as the input to the system which further reduces the computations. Four deep learning models are proposed to extract the most discriminative features which enhance the classification accuracy and prediction time. The proposed approach takes advantage of the convolutional neural network in extracting the significant spatial features from different scalp positions and the recurrent neural network in expecting the incidence of seizures earlier than the current methods. A semi-supervised approach based on transfer learning technique is introduced to improve the optimization problem. A channel selection algorithm is proposed to select the most relevant EEG channels which makes the proposed system good candidate for real-time usage. An effective test method is utilized to ensure robustness. The achieved highest accuracy of 99.6% and lowest false alarm rate of 0.004 h - 1 along with very early seizure prediction time of 1 h make the proposed method the most efficient among the state of the art.