EEG Database Research Articles

Automated Rest EEG-Based Diagnosis of Depression and Schizophrenia Using a Deep Convolutional Neural Network

Depression (DP) and schizophrenia (SCZ) are both highly prevalent psychiatric disorders, and their diagnosis depends on the examination of symptoms and clinical tests, which can be subjective. As a measure of real-time neural activity, Electroencephalographic (EEG) has shown its usability to classify people either as normal or as having DP or SCZ, but automatic classification between the three categories (DP, SCZ and the normal) was rarely reported. Here, we propose an automatic diagnostic framework based on a convolutional neural network called the Multi-Channel Frequency Network (MUCHf-Net), which automatically learns feature representations of EEGs that characterize them as normal, DP, or SCZ. Two EEG databases were used in this study, the first one contains EEGs from 300 individuals (DP: 100, SCZ: 100, normal: 100) collecting from our hospital, and the second contains EEGs from 30 individuals (DP: 10, SCZ: 10, normal: 10) from public available datasets, and the spectrum matrices from these multi-channel EEGs were feed into MUCHf-Net. The results showed that: (1) MUCHf-Net accurately distinguished normal EEGs from DP or SCZ EEGs (accuracy: 91.12%; F1 score: 0.8947); (2) low-frequency bands (delta, theta, alpha) contributed the most important information to the classification model; (3) features located in the frontal and parietal lobes contributed more than other regions did; (4) MUCHf-Net fine-tuned on public datasets also had high classification accuracy: 87.71% (triple: normal, SCZ or DP) and 79.27% (binary: psychiatric disorders (DP or SCZ) or normal). Our study shows that deep leaning has the potential to become an important tool for assisting in the diagnosis of psychiatric disorders.

A combination of statistical parameters for epileptic seizure detection and classification using VMD and NLTWSVM

The epileptic seizure detection and classification is of great significance for clinical diagnosis and treatment. To realize the detection and classification of epileptic seizure, this paper proposes a method based on the combination of signal decomposition and statistical methods. First, the algorithm of variational mode decomposition (VMD) is applied to extract the components of intrinsic mode functions (IMFs) by decomposing the EEG signals. Then the statistical method is utilized to calculate the eight features of maximum, minimum, average, variance, skewness, kurtosis, coefficient of variation and volatility index for each extracted IMF component. Finally, the best combinations of extracted features are fed into the non-linear twin support vector machine (NLTWSVM) to classify the epileptic signals. The EEG database from University of Bonn is used to confirm the effectiveness of the proposed method for epileptic seizure detection. The final experimental results demonstrate that the classification accuracy can reach 98.86%, 98.37%, 99.02%, 99.41% and 99.57% for the database of C-E, D-E, CD-E, ABCD-E and AB-CD-E, respectively. The TUSZ corpus in the TUH EEG corpus is also used to classify epileptic seizure types using the method in this article. The result is expressed by the confusion matrix and the weighted F1 score is 0.923, which shows this method has potential to help experienced neurophysiologists classify epileptic seizure types in the clinic.

Siamese Network‐Based Feature Transformation for Improved Automated Epileptic Seizure Detection

Epilepsy is a common electrophysiological disorder of the brain, detected mainly by electroencephalogram (EEG) signals. Since correctly diagnosing epilepsy seizures by monitoring the EEG signal is very tedious and time‐consuming for a neurologist, a growing number of studies have been conducted on developing automated epileptic seizure detection (AESD). In this work, a novel supervised model is proposed for AESD. Initially, the EEG signals are collected from Bonn University EEG (BU‐EEG) database. Then, empirical mode decomposition and feature extraction (combination of entropy, frequency, and statistical features) are applied to extract the features. Furthermore, Siamese network is utilized to lessen the number of extracted features and obtain the most discriminative features. Then, these features are exploited to classify seizure and non‐seizure EEG signals by using a support vector machine classifier. This paper examines the Siamese network’s contribution as a learning‐based feature transformation in improving seizure detection performance. The numerical results confirm that the proposed framework can achieve a perfect classification performance (100% accuracy). This approach can constructively help doctors to detect epileptic seizure activity and reduce their workload.

Normal Inverse Gaussian Features for EEG-Based Automatic Emotion Recognition

Electroencephalography (EEG)-based emotion recognition is crucial in the domain of Human-Computer Interaction (HCI), which gained significant attention in recent years. However, the non-stationarity and chaotic nature of the EEG signals pose challenges and restrict the state-of-the-art techniques from precisely identifying distinct emotional states from the EEG data and hence offer limited emotion recognition performance. To capture the underlying non-linear characteristics of EEG, this study employed a novel local mean decomposition (LMD) algorithm which decomposes EEG signals into product functions (PFs). Further PFs are modeled by Normal Inverse Gaussian (NIG) probability density function (PDF) parameters. Thus these PDF features are fed to an optimized Adaboost classifier developed with the help of a cross-validation approach. The novelty of the work lies in the NIG modeling of LMD domain PFs to identify specific emotions from the EEG signals. The significance of the NIG parameters is illustrated by qualitative, pictorial, and statistical analyses. To assess the efficiency of the proposed approach, intensive experiments are conducted on open-source datasets, SJTU Emotion EEG Dataset (SEED), SEED-IV, and Database for Emotion Analysis of Physiological Signals (DEAP). The emotion recognition performance is evaluated in terms of heat maps, receiver operating characteristics (ROC), and accuracy. The proposed emotion recognition system outperformed the state-of-art methods and achieved a maximum accuracy of 97.3%, 98%, and 98.6% with the cross-validation approach and 93.23%, 94.87%, and 95.58% for the cross-subject validation approach using DEAP and SEED, SEED-IV datasets, respectively.

Seizure Detection Based on Improved Genetic Algorithm Optimized Multilayer Network

With the increasment of epilepsy patients, traditional epileptic seizure recognition is generally completed by encephalography (EEG) technicians, which is time-consuming and labor-intensive, so the automatic detection of seizure is imminent. This paper proposes a method which constructs a multi-layer network and extracts the same features in each network optimized by improved genetic algorithm (IGA). Among them, the multi-layer network refers to the three-layer network constructed by pearson correlation coefficient, mutual information and permutation disalignment index respectively. There is a lack of research on the fusion and comparison of different networks in previous studies. Therefore, this paper analyzes the effectiveness of different networks by studying the fusion relationship of different networks, and further uses IGA for iterative optimization with constraints to weight the network and features, and finally uses the random forest classifier to automatically detect epileptic seizures. On CHB-MIT database, accuracy (ACC), specificity (SPE), sensitivity (SEN) and F1 score (F1) of the method proposed in this paper reach 97.26%, 97.55%, 96.89% and 97.11%, respectively. On Siena scalp EEG database, ACC, SPE, SEN and F1 reach 98.88%, 99.13%, 98.36% and 98.75%, respectively. The results show that the joint detection effect of the multi-layer network is better than the combined effect of other networks, and IGA can improve the effect of seizure detection.

A Review on Machine Learning Approaches in Identification of Pediatric Epilepsy.

Epilepsy is the second most common neurological disease after Alzheimer. It is a disorder of the brain which results in recurrent seizures. Though the epilepsy in general is considered as a serious disorder, its effects in children are rather dangerous. It is mainly because it reasons a slower rate of development and a failure to improve certain skills among such children. Seizures are the most common symptom of epilepsy. As a regular medical procedure, the specialists record brain activity using an electroencephalogram (EEG) to observe epileptic seizures. The detection of these seizures is performed by specialists, but the results might not be accurate and depend on the specialist’s experience; therefore, automated detection of epileptic pediatric seizures might be an optimal solution. In this regard, several techniques have been investigated in the literature. This research aims to review the approaches to pediatric epilepsy seizures’ identification especially those based on machine learning, in addition to the techniques applied on the CHB-MIT scalp EEG database of epileptic pediatric signals.

A hybrid technique for EEG signals evaluation and classification as a step towards to neurological and cerebral disorders diagnosis

Electroencephalography (EEG) signals are commonly used to identify and diagnose brain disorders. Each EEG normal waveform consists of the following waveforms: Gamma(γ) wave, Beta (β) wave, Alpha (α), Theta (θ), and Delta (δ). The term Neurological Diseases ” NurDis ” is used to describe a variety of conditions that affect the nervous Epilepsy, neuro infections (bacterial and viral), brain tumors, cerebrovascular diseases, Alzheimer’s disease, and various dementias are all examples of neurological disorders. Encephalitis is one of the illnesses that affects the brain. The EEG signals used in this paper were from the CHB-MIT Scalp EEG database. The discrete wavelet transform (DWT) was utilized to extract characteristics from the filtered EEG data. Finally, classifiers such as K Nearest Neighbor (KNN) and Support vector machine (SVM) were used to categorize the EEG signals into normal and pathological signal classes using all of the computed characteristics. In order to categorize the signal in a normal and anomalous group, the KNN and SVM classifiers are employed. For both classifiers, performance assessments (accuracy, sensitivity and specificity) are determined. KNN classifier accuracy is 71.88%, whereas SVM classifier accuracy is 81.23%. The sensitivity of KNN and SVM are 80.14% and 77.31%, respectively. The KNN classification specificity is 69.62% and the SVM classification specificity is 98%. Both classifiers performance is evaluated using the confusion matrix.

Epileptic Seizure Detection Based on Bidirectional Gated Recurrent Unit Network

Visual inspection of long-term electroencephalography (EEG) is a tedious task for physicians in neurology. Based on bidirectional gated recurrent unit (Bi-GRU) neural network, an automatic seizure detection method is proposed in this paper to facilitate the diagnosis and treatment of epilepsy. Firstly, wavelet transforms are applied to EEG recordings for filtering pre-processing. Then the relative energies of signals in several particular frequency bands are calculated and inputted into Bi-GRU network. Afterwards, the outputs of Bi-GRU network are further processed by moving average filtering, threshold comparison and seizure merging to generate the discriminant results that the tested EEG belong to seizure or not. Evaluated on CHB-MIT scalp EEG database, the proposed seizure detection method obtained an average sensitivity of 93.89% and an average specificity of 98.49%. 124 out of 128 seizures were correctly detected and the achieved average false detection rate was 0.31 per hour on 867.14 h testing data. The results show the superiority of Bi-GRU network in seizure detection and the proposed detection method has a promising potential in the monitoring of long-term EEG.

Automatic seizure detection with different time delays using SDFT and time-domain feature extraction.

Automatic seizure detection is important for fast detection of the seizure because the way that the expert denotes and searches for seizure in the long signal takes time. The most common way to detect seizures automatically is to use an electroencephalogram (EEG). Many studies have used feature extraction that needs time for calculation. In this study, sliding discrete Fourier transform (SDFT) was applied for conversion to a frequency domain without using a window, which was compared with using window for feature selection. SDFT was calculated for each time series sample directly without any delay by using a simple infinite impulse response (IIR) structure. The EEG database of Bonn University was used to test the proposed method, and two cases were defined to examine a two-classifier feedforward neural network and an adaptive network-based fuzzy inference system. Results revealed that the maximum accuracies were 93% without delay and 99.8% with a one-second delay. This delay accrued because the average was taken for the results with a one-second window.

Peri‐ictal and non‐seizure EEG event detection using generated metadata

AbstractLack of open access, seizure specific database has hindered the development of Automated Seizure Detection System (ASDS) along with state‐of‐the‐art feature selection and classification methods. Available databases contain noise, artefacts, different length & time EEG segments, associated comorbidities, clinical settings, and epilepsy/seizure types etc. Pre‐processing of such continuous EEG segments requires significant amount of time and may feed redundant information (with reference to a seizure event) to the classification model leading to inaccurate, real‐time seizure event detection systems. Hence, this paper proposes a metadata generation of large EEG databases (here, CHB‐MIT EEG scalp database v.1.0.0) for ASDS. We elucidate the need to generate seizure sensitive data through peri‐ictal and non‐seizure EEG segments. This paper performs multi‐variate analysis and two class (non‐seizure and seizure event) classification between these fixed length and time EEG segments from support vector machine (SVM) and k‐NN classifier. We thoroughly analysed the variation and dependence of different kernels, cost function, gamma, and degree for SVM based pipeline. The proposed pipeline has been compared with state‐of‐the‐art pipelines and has achieved good classification score. Such methods will help in development of generalised approach toward handling large EEG databases and machine learning applications for seizure detection and prediction.

Electroencephalography Correlates of Well-Being Using a Low-Cost Wearable System.

Electroencephalography (EEG) alpha asymmetry is thought to reflect crucial brain processes underlying executive control, motivation, and affect. It has been widely used in psychopathology and, more recently, in novel neuromodulation studies. However, inconsistencies remain in the field due to the lack of consensus in methodological approaches employed and the recurrent use of small samples. Wearable technologies ease the collection of large and diversified EEG datasets that better reflect the general population, allow longitudinal monitoring of individuals, and facilitate real-world experience sampling. We tested the feasibility of using a low-cost wearable headset to collect a relatively large EEG database (N = 230, 22–80 years old, 64.3% female), and an open-source automatic method to preprocess it. We then examined associations between well-being levels and the alpha center of gravity (CoG) as well as trait EEG asymmetries, in the frontal and temporoparietal (TP) areas. Robust linear regression models did not reveal an association between well-being and alpha (8–13 Hz) asymmetry in the frontal regions, nor with the CoG. However, well-being was associated with alpha asymmetry in the TP areas (i.e., corresponding to relatively less left than right TP cortical activity as well-being levels increased). This effect was driven by oscillatory activity in lower alpha frequencies (8–10.5 Hz), reinforcing the importance of dissociating sub-components of the alpha band when investigating alpha asymmetries. Age was correlated with both well-being and alpha asymmetry scores, but gender was not. Finally, EEG asymmetries in the other frequency bands were not associated with well-being, supporting the specific role of alpha asymmetries with the brain mechanisms underlying well-being levels. Interpretations, limitations, and recommendations for future studies are discussed. This paper presents novel methodological, experimental, and theoretical findings that help advance human neurophysiological monitoring techniques using wearable neurotechnologies and increase the feasibility of their implementation into real-world applications.

Circannual incidence of seizure evacuations from the Canadian Arctic

ObjectivesEmerging evidence suggests that circadian rhythms affect seizure propensity in addition to, and possibly independent of, sleep-wake states. Subject to extreme seasonal changes in light and dark, the northerly Arctic can serve as a “natural experiment” to assess the real-life impact of environmental influences on seizure severity. Therefore, we evaluated the timing of seizure evacuations over 11.25 years in a well-defined region of the Canadian Arctic. MethodsRetrospective review of EEG database and patient records at the single “bottleneck” hospital to which all patients from the Kivalliq Region in Nunavut, Canada are evacuated for seizure emergencies. We calculated the mean resultant length (MRL) of circular data for circannual analysis, and conducted Rayleigh’s test to assess for a statistical departure from circular uniformity. ResultsScreening 40,392 EEGs, we found 117 medical evacuations from 99 distinct individuals from September 2009 to November 2020. Most evacuations occurred month-wise in May (19%); week-wise within a 7-day period in February (5%), June (5%), or November (5%); and day-wise within a 24-hour period in June (3%) or November (3%). Maximal MRL clustering occurred in April no matter if analyzed by day (0.16333, p = 0.04), week (0.16296, p = 0.04), or month (0.1736, p = 0.03). ConclusionsA relative circannual increase in seizure evacuations between the winter and summer solstices may be related to increasing sleep loss when day length grows. Fewer evacuations between the summer and winter solstices may be related to decreased daylight and “catching up” on sleep when night length grows. Additional factors likely also play a role in circannual variation of seizure evacuations in the Arctic, which warrants further research.

Generative adversarial network and convolutional neural network-based EEG imbalanced classification model for seizure detection

Automatic seizure detection technology is of great significance to reduce workloads of neurologists for epilepsy diagnosis and treatments. Imbalanced classification is a challenge in seizure detection from long-term continuous EEG recordings, as the durations of the seizure events are much shorter than the non-seizure periods. An imbalanced deep learning model is proposed in this paper to improve the performance of seizure detection. To modify imbalanced EEG data distribution, a generative adversarial network (GAN) that is a strong candidate for data enhancement is built to produce the seizure-period EEG data used for forming a more balanced training set. Next, a pyramidal one-dimensional convolutional neural network (1DCNN) is designed to deal with 1D EEG signals and trained on the augmented training set that consists of both original and generated EEG data. Compared to the conventional 2DCNNs, the deep architecture of the 1DCNN reduces the training parameters so as to greatly increase the training speed. The proposed method is evaluated on three publicly available EEG databases. After data augmentation by the GAN, the designed 1DCNN shows much better classification for seizure detection, achieving competitive results over the three EEG databases, which demonstrates the generalizability of this method across different databases. Comparison with other published methods indicates its enhanced detection performance for imbalanced EEG data.

Anterior temporal lobectomy: A cross-sectional observational study of potential surgical candidates at a single institute.

Background: Epilepsy is a common neurological disorder, associated with serious cognitive, physical, and psychosocial burdens. Mesial temporal lobe epilepsy (mTLE) is the commonest form of focal epilepsy. The aim of this study was to establish the incidence of patients with electroencephalographic epileptiform discharges consistent with mTLE attending a tertiary hospital in South Africa, and determine whether these patients may be candidates for anterior temporal lobectomy.Methods: This was a cross-sectional observational study of all patients receiving scalp electroencephalograms (EEG) performed at the Groote Schuur Hospital Neurophysiology laboratory during the period January 1, 2017–December 31, 2019. Where magnetic resonance imaging (MRI) brain scans had been performed, these were assessed for corroborative evidence of mTLE.Results: Over the 3-year period, 4 342 EEGs were assessed. A total of 411 (11%) showed epileptiform discharges consistent with all epilepsy types. Of these, 327 (69%) were of focal onset and 108 (33% of all focal onset epilepsies) were consistent with mTLE. Of the patients with electroencephalographic features of mTLE, only 27 (25%) had had MRI brain scans performed according to an epilepsy surgery protocol. None of these patients had been considered for surgery.Conclusion: Surgery, especially anterior temporal lobectomy, is widely acknowledged to be an efficacious and cost-effective intervention in patients with drug-resistant mTLE. The findings of our study suggest that patients with mTLE in our setting are under-investigated for potential surgery; and that it is under-utilized. These findings are in line with similar studies in both well-resourced and resource-constrained countries. Our study also highlights the utility of EEG as a practical screening tool to identify potential surgical candidates, as well as the establishment of an EEG and MRI database to assist in recognizing these patients.

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.

Brain fragility among middle-aged and elderly patients from electroencephalogram during induction of anaesthesia.

Brain fragility among middle-aged and elderly patients from electroencephalogram during induction of anaesthesia.

Seizure detection from multi-channel EEG using entropy-based dynamic graph embedding

An epileptic seizure is a chronic disease with sudden abnormal discharge of brain neurons, which leads to transient brain dysfunction. To detect epileptic seizures, we propose a novel idea based on a dynamic graph embedding model. The dynamic graph is built by identifying the correlation among the multi-channel EEG signals. Graph entropy measurement is exploited to calculate the similarity among the graph at each time interval and construct the graph embedding space. Since the abnormal electrical brain activity causes the epileptic seizure, the graph entropy during the seizure time interval is different from other time intervals. Therefore, we propose an entropy-based dynamic graph embedding model to cluster the graphs, and the graphs with epileptic seizures are discriminated. We applied the proposed approach to the Children Hospital Boston-Massachusetts Institute of Technology Scalp EEG database. The results have shown that the proposed approach outperformed the baselines by 1.4% with respect to accuracy.

Basic study of epileptic seizure detection using a single-channel frontal EEG and a pre-trained ResNet.

Epilepsy is a neurological disorder that causes sudden seizures due to abnormal excitation of neurons in the brain. Approximately 30 % of patients cannot control their seizures using medication. In addition, since seizures can occur anywhere and at any time, caregivers must always be with the patient. Various researchers have developed seizure detection methods using multichannel EEG to improve the quality of life of patients and caregivers. However, the large size of the measurement device impedes transportation. We believe that a portable measurement device with a small number of channels is suitable for detecting seizures in daily life. Therefore, we need a system that can detect seizures using a small number of channels. The purpose of this research is to develop a seizure detection algorithm using a single-channel frontal EEG and to confirm its basic performance. We used EEG signals from a single electrode position (Fp1-F7, Fp2-F8), which is a bipolar derivation of the frontal region. We segmented the EEG using a 2 s sliding window with 50 % overlap and converted the segments into images. After preprocessing, we fine-tuned ResNet18, pre-trained on ImageNet, and developed an ensemble classification method. In the experiments with 10 epileptic patients (3 - 19 years old) registered in the CHB-MIT scalp EEG database, the results showed that the average sensitivity was 88.73 %, the average specificity was 98.98 %, and the average detection latency time was 7.39 s. In conclusion, the developed algorithm was validated as sufficiently accurate to detect epileptic seizures.Clinical Relevance- This establishes an image recognition algorithm that can detect epileptic seizures using a single- channel frontal EEG.

Altered EEG Brain Networks in Patients with Acute Peripheral Herpes Zoster.

ObjectiveTo investigate whether the brain networks changed in patients with acute peripheral herpes zoster (HZ).MethodsWe reviewed the EEG database in Jianyang People’s Hospital. Patients with acute HZ (n=71) were enrolled from January 2016 to December 2020. Each included subject underwent a ten-minute and 16-channel EEG examination. Five epochs of 10-second EEG data in resting-state were collected from each HZ patient. Five 10-second resting-state EEG epochs from sex- and age-matched healthy controls (HC, n=71) who reported no history of neurological or psychiatric disorders and visited the hospital for routine physical examinations were collected. Brain network and graph theory analysis based on phase locking value parameter and functional ICA were performed using a self-writing Matlab code and the LORETA KEY tool.ResultsCompared with the HC group, the HZ patients showed significant altered brain networks. The graph theory analysis revealed that the clustering coefficient and local efficiency of full band in HZ patients were lower than those in HC group (P<0.05). In beta band, the global efficiency and local efficiency of HZ patients group decreased, compared with healthy group (P<0.05). The functional ICA showed that three components showed significant differences between the two groups. In component 2, HZ patients showed excess superior frontal gyrus (BA10) neuro oscillation in delta band and less medial frontal gyrus (BA 11) neuro oscillation in beta and gamma bands than that in HCs. And for component 3, the alpha band of the HZ patients presented increased neuro activities in superior frontal gyrus (BA 11) and decreased neuro activities in occipital lobe (BA 18). In component 4, the inferior frontal gyrus (BA 47) showed excess activity in the left hemisphere and reduced activity in the right hemisphere in delta band, compared with HC group.ConclusionAltered brain networks exist in resting-state EEG data of patients with acute HZ. The changes of EEG brain networks in HZ patients are characterized by decreased global efficiency and local efficiency in beta band. Moreover, the spontaneous oscillation of some brain regions involving pain management and the connectivity of default mode network changed in HZ patients. Our study provided novel understanding of HZ from an electrophysiological view, and led to converging evidence for treatment of HZ with neural regulation in future.

Classifying Single Channel Epileptic EEG data based on Sparse Representation using Shallow Autoencoder.

Patient independent epileptic seizure detection algorithm for scalp electroencephalogram (EEG) data is pro- posed in this paper. Principal motivation of this work is to integrate neural and conventional machine learning methods to develop a classification system which can advance the current wearable health systems in terms of computational complexity and accuracy. Being based on processing a single channel EEG processing, the approach is suitable for usage with small wireless sensors. A shallow autoencoder model is utilized for sparse representation of the EEG signal followed by k-nearest neighbor (kNN) classifier to categorize the data as epileptic or non-epileptic. Using a single EEG channel an optimum sparsity level is explored in the encoded sample. Attaining an accuracy, sensitivity and specificity of 98.85%, 99.29% and 98.86% respectively, for CHB-MIT scalp EEG database, proposed classification method outperforms state of- the-art seizure detection methodologies. Experiments has shown that this performance was possible by using a sparsity level of 4 in the auto-encoder. Furthermore, use of shallow learning instead of deep learning approach for generation of sparse but effective representation is computationally lighter than many other feature extraction and preprocessing methods.