scholarly journals Cross-Subject Seizure Detection in EEGs Using Deep Transfer Learning

2020 ◽  
Vol 2020 ◽  
pp. 1-8
Author(s):  
Baocan Zhang ◽  
Wennan Wang ◽  
Yutian Xiao ◽  
Shixiao Xiao ◽  
Shuaichen Chen ◽  
...  
Keyword(s):  
Binary Classification ◽  
Seizure Detection ◽  
Model Parameters ◽  
Time Frequency ◽  
Scalp Eeg ◽  
Original Dataset ◽  
Input Dataset ◽  
Eeg Recordings ◽  
Short Time ◽  
Patients With Epilepsy

Electroencephalography (EEG) plays an import role in monitoring the brain activities of patients with epilepsy and has been extensively used to diagnose epilepsy. Clinically reading tens or even hundreds of hours of EEG recordings is very time consuming. Therefore, automatic detection of seizure is of great importance. But the huge diversity of EEG signals belonging to different patients makes the task of seizure detection much challenging, for both human experts and automation methods. We propose three deep transfer convolutional neural networks (CNN) for automatic cross-subject seizure detection, based on VGG16, VGG19, and ResNet50, respectively. The original dataset is the CHB-MIT scalp EEG dataset. We use short time Fourier transform to generate time-frequency spectrum images as the input dataset, while positive samples are augmented due to the infrequent nature of seizure. The model parameters pretrained on ImageNet are transferred to our models. And the fine-tuned top layers, with an output layer of two neurons for binary classification (seizure or nonseizure), are trained from scratch. Then, the input dataset are randomly shuffled and divided into three partitions for training, validating, and testing the deep transfer CNNs, respectively. The average accuracies achieved by the deep transfer CNNs based on VGG16, VGG19, and ResNet50 are 97.75%, 98.26%, and 96.17% correspondingly. On those results of experiments, our method could prove to be an effective method for cross-subject seizure detection.

scholarly journals Zero-crossing patterns reveal subtle epileptiform discharges in the scalp EEG

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Jan Pyrzowski ◽  
Jean- Eudes Le Douget ◽  
Amal Fouad ◽  
Mariusz Siemiński ◽  
Joanna Jędrzejczak ◽  
...  
Keyword(s):  
Normal Subjects ◽  
Intracranial Eeg ◽  
Zero Crossing ◽  
Temporal Precision ◽  
Scalp Eeg ◽  
Epileptiform Discharges ◽  
Interictal Epileptiform Discharges ◽  
Eeg Recordings ◽  
Patients With Epilepsy ◽  
Low Amplitude

AbstractClinical diagnosis of epilepsy depends heavily on the detection of interictal epileptiform discharges (IEDs) from scalp electroencephalographic (EEG) signals, which by purely visual means is far from straightforward. Here, we introduce a simple signal analysis procedure based on scalp EEG zero-crossing patterns which can extract the spatiotemporal structure of scalp voltage fluctuations. We analyzed simultaneous scalp and intracranial EEG recordings from patients with pharmacoresistant temporal lobe epilepsy. Our data show that a large proportion of intracranial IEDs manifest only as subtle, low-amplitude waveforms below scalp EEG background and could, therefore, not be detected visually. We found that scalp zero-crossing patterns allow detection of these intracranial IEDs on a single-trial level with millisecond temporal precision and including some mesial temporal discharges that do not propagate to the neocortex. Applied to an independent dataset, our method discriminated accurately between patients with epilepsy and normal subjects, confirming its practical applicability.

scholarly journals High similarity between EEG from subcutaneous and proximate scalp electrodes in patients with temporal lobe epilepsy

2018 ◽  
Vol 120 (3) ◽  
pp. 1451-1460 ◽  
Author(s):  
Sigge Weisdorf ◽  
Sirin W. Gangstad ◽  
Jonas Duun-Henriksen ◽  
Karina S. S. Mosholt ◽  
Troels W. Kjær
Keyword(s):  
Temporal Lobe Epilepsy ◽  
Temporal Lobe ◽  
Temporal Dynamics ◽  
Real Life ◽  
High Similarity ◽  
Time Frequency ◽  
Scalp Eeg ◽  
Eeg Recordings ◽  
The Impact

Subcutaneous recording using electroencephalography (EEG) has the potential to enable ultra-long-term epilepsy monitoring in real-life conditions because it allows the patient increased mobility and discreteness. This study is the first to compare physiological and epileptiform EEG signals from subcutaneous and scalp EEG recordings in epilepsy patients. Four patients with probable or definite temporal lobe epilepsy were monitored with simultaneous scalp and subcutaneous EEG recordings. EEG recordings were compared by correlation and time-frequency analysis across an array of clinically relevant waveforms and patterns. We found high similarity between the subcutaneous EEG channels and nearby temporal scalp channels for most investigated electroencephalographic events. In particular, the temporal dynamics of one typical temporal lobe seizure in one patient were similar in scalp and subcutaneous recordings in regard to frequency distribution and morphology. Signal similarity is strongly related to the distance between the subcutaneous and scalp electrodes. On the basis of these limited data, we conclude that subcutaneous EEG recordings are very similar to scalp recordings in both time and time-frequency domains, if the distance between them is small. As many electroencephalographic events are local/regional, the positioning of the subcutaneous electrodes should be considered carefully to reflect the relevant clinical question. The impact of implantation depth of the subcutaneous electrode on recording quality should be investigated further. NEW & NOTEWORTHY This study is the first publication comparing the detection of clinically relevant, pathological EEG features from a subcutaneous recording system designed for out-patient ultra-long-term use to gold standard scalp EEG recordings. Our study shows that subcutaneous channels are very similar to comparable scalp channels, but also point out some issues yet to be resolved.

Detection and Removal of Muscle Artifacts from Scalp EEG Recordings in Patients with Epilepsy

2014 ◽  
Author(s):  
Maria Anastasiadou ◽  
Avgis Hadjipapas ◽  
Manolis Christodoulakis ◽  
Eleftherios S. Papathanasiou ◽  
Savvas S. Papacostas ◽  
...  
Keyword(s):  
Scalp Eeg ◽  
Eeg Recordings ◽  
Patients With Epilepsy

scholarly journals Automatic detection of generalized paroxysmal fast activity in Lennox-Gastaut syndrome using a bi-modal EEG time-frequency feature

2020 ◽  
Author(s):  
Amir Omidvarnia ◽  
Aaron E.L. Warren ◽  
Linda J. Dalic ◽  
Mangor Pedersen ◽  
John S. Archer ◽  
...  
Keyword(s):  
Automatic Detection ◽  
Time Frequency ◽  
Scalp Eeg ◽  
Epileptiform Discharges ◽  
Fast Activity ◽  
Brain Maps ◽  
Detection Approach ◽  
Interictal Eeg ◽  
Eeg Recordings ◽  
Frequency Feature

AbstractObjectiveMark-up of generalized interictal epileptiform discharges (IEDs) on EEG is an important step in the diagnosis and characterization of epilepsy. However, manual EEG mark-up is a time-consuming, subjective, and highly specialized task where the human reviewer needs to visually inspect a large amount of data to facilitate accurate clinical decisions. The objective of this study was to develop a framework for automated detection of generalized paroxysmal fast activity (GPFA), which is a characteristic type of generalized IED seen in scalp EEG recordings of patients with Lennox-Gastaut syndrome (LGS), a severe form of drug-resistant generalized epilepsy.MethodsWe studied 13 children with LGS who had GPFA events in their interictal EEG recordings. Time-frequency information derived from manually marked IEDs across multiple EEG channels was used to automatically detect similar events in each patient’s interictal EEG. We validated true positives and false positives of the proposed spike detection approach using both standalone scalp EEG and simultaneous EEG-functional MRI (EEG-fMRI) recordings.ResultsGPFA events displayed a consistent low-high frequency arrangement in the time-frequency domain. This ‘bi-modal’ spectral feature was most prominent over frontal EEG channels. Our automatic detection approach using this feature identified likely epileptic events with similar time-frequency properties to the manually marked GPFAs. Brain maps of EEG-fMRI signal change during these automatically detected IEDs were comparable to the EEG-fMRI brain maps derived from manual IED mark-up.ConclusionGPFA events have a characteristic bi-modal time-frequency feature that can be automatically detected from scalp EEG recordings in patients with LGS. Validity of this time-frequency feature is demonstrated by EEG-fMRI analysis of automatically detected events, which recapitulates the brain maps we have previously shown to underlie generalized IEDs in LGS.SignificanceThis study provides a novel methodology that paves the way for quick, automated, and objective inspection of generalized IEDs in LGS. The proposed framework may be extendable to a wider range of epilepsy syndromes in which monitoring the burden of epileptic activity can aid clinical decision-making. For example, automated quantification of generalized discharges may permit faster assessment of treatment response and estimation of future seizure risk.

scholarly journals A Comparison of Deep Neural Networks for Seizure Detection in EEG Signals

2019 ◽  
Author(s):  
Poomipat Boonyakitanont ◽  
Apiwat Lek-uthai ◽  
Krisnachai Chomtho ◽  
Jitkomut Songsiri
Keyword(s):  
Neural Networks ◽  
Seizure Detection ◽  
Machine Learning Techniques ◽  
Diagnostic Process ◽  
Eeg Signals ◽  
Time Frequency ◽  
Scalp Eeg ◽  
Learning Techniques ◽  
Frequency Domains ◽  
2D Filter

AbstractThis paper aims to apply machine learning techniques to an automated epileptic seizure detection using EEG signals to help neurologists in a time-consuming diagnostic process. We employ two approaches based on convolution neural networks (CNNs) and artificial neural networks (ANNs) to provide a probability of seizure occurrence in a windowed EEG recording of 18 channels. In order to extract relevant features based on time, frequency, and time-frequency domains for these networks, we consider an improvement of the Bayesian error rate from a baseline. Features of which the improvement rates are higher than the significant level are considered. These dominant features extracted from all EEG channels are concatenated as the input for ANN with 7 hidden layers, while the input of CNN is taken as raw multi-channel EEG signals. Using multi-concept of deep CNN in image processing, we exploit 2D-filter decomposition to handle the signal in spatial and temporal domains. Our experiments based on CHB-MIT Scalp EEG Database showed that both ANN and CNN were able to perform with the overall accuracy of up to 99.07% and F1-score of up to 77.04%. ANN with dominant features is more capable of detecting seizure events than CNN whereas CNN requiring no feature extraction is slightly better than ANN in classification accuracy.

Prediction and detection of seizures from simultaneous thalamic and scalp electroencephalography recordings

2016 ◽  
Vol 126 (6) ◽  
pp. 2036-2044 ◽  
Author(s):  
Rosa Q. So ◽  
Vibhor Krishna ◽  
Nicolas Kon Kam King ◽  
Huijuan Yang ◽  
Zhuo Zhang ◽  
...  
Keyword(s):  
False Alarm ◽  
False Alarm Rate ◽  
Detection Rate ◽  
Roc Curves ◽  
Seizure Detection ◽  
Treatment Of Epilepsy ◽  
Scalp Eeg ◽  
Prediction Rate ◽  
Eeg Recordings ◽  
Using Data

OBJECTIVEThe authors explored the feasibility of seizure detection and prediction using signals recorded from the anterior thalamic nucleus, a major target for deep brain stimulation (DBS) in the treatment of epilepsy.METHODSUsing data from 5 patients (13 seizures in total), the authors performed a feasibility study and analyzed the performance of a seizure prediction and detection algorithm applied to simultaneously acquired scalp and thalamic electroencephalography (EEG). The thalamic signal was obtained from DBS electrodes. The applied algorithm used the similarity index as a nonlinear measure for seizure identification, with patient-specific channel and threshold selection. Receiver operating characteristic (ROC) curves were calculated using data from all patients and channels to compare the performance between DBS and EEG recordings.RESULTSThalamic DBS recordings were associated with a mean prediction rate of 84%, detection rate of 97%, and false-alarm rate of 0.79/hr. In comparison, scalp EEG recordings were associated with a mean prediction rate of 71%, detection rate of 100%, and false-alarm rate of 1.01/hr. From the ROC curves, when considering all channels, DBS outperformed EEG for both detection and prediction of seizures.CONCLUSIONSThis is the first study to compare automated seizure detection and prediction from simultaneous thalamic and scalp EEG recordings. The authors have demonstrated that signals recorded from DBS leads are more robust than EEG recordings and can be used to predict and detect seizures. These results indicate feasibility for future designs of closed-loop anterior nucleus DBS systems for the treatment of epilepsy.

scholarly journals Time-Varying EEG Correlations Improve Automated Neonatal Seizure Detection

2019 ◽  
Vol 29 (04) ◽  
pp. 1850030 ◽  
Author(s):  
Karoliina T. Tapani ◽  
Sampsa Vanhatalo ◽  
Nathan J. Stevenson
Keyword(s):  
State Of The Art ◽  
Seizure Detection ◽  
Detection Algorithm ◽  
Neonatal Seizure ◽  
Support Vector ◽  
Time Varying ◽  
Term Neonates ◽  
Time Frequency ◽  
Correlation Measures ◽  
Eeg Recordings

The aim of this study was to develop methods for detecting the nonstationary periodic characteristics of neonatal electroencephalographic (EEG) seizures by adapting estimates of the correlation both in the time (spike correlation; SC) and time–frequency domain (time–frequency correlation; TFC). These measures were incorporated into a seizure detection algorithm (SDA) based on a support vector machine to detect periods of seizure and nonseizure. The performance of these nonstationary correlation measures was evaluated using EEG recordings from 79 term neonates annotated by three human experts. The proposed measures were highly discriminative for seizure detection (median [Formula: see text]: 0.933 IQR: 0.821–0.975, median [Formula: see text]: 0.883 IQR: 0.707–0.931). The resultant SDA applied to multi-channel recordings had a median AUC of 0.988 (IQR: 0.931–0.998) when compared to consensus annotations, outperformed two state-of-the-art SDAs [Formula: see text] and was noninferior to the human expert for 73/79 of neonates.

scholarly journals ROC Analysis of EEG Subbands for Epileptic Seizure Detection using Naïve Bayes Classifier

2021 ◽  
Author(s):  
Mustafa Sameer ◽  
Bharat Gupta
Keyword(s):  
Epileptic Seizure ◽  
Roc Analysis ◽  
Naive Bayes ◽  
Seizure Detection ◽  
Naïve Bayes ◽  
Beta Band ◽  
Epileptic Seizure Detection ◽  
Time Frequency ◽  
Image Descriptors ◽  
Short Time

This paper presents analysis of Electroencephalograms (EEGs) and subbands (delta, theta, alpha, beta, gamma) using image descriptors for epileptic seizure detection. Short-time Fourier transform (STFT) has been utilized to convert 1-D EEG data into image. All subbands are separated from the time-frequency (t-f) matrix and Haralick features of each subband is fed in the Naïve Bayes (NB) classifier. Receiver operating characteristic (ROC) analysis has been used for performance evaluation of classifier. Among all subbands, gamma band alone shows a maximum AUC of 0.98 to classify between ictal and healthy class, while beta band shows a maximum AUC of 0.96 to differentiate between ictal and interictal class. Significance of this work is it shows the medical advantage of different subbands for the detection process.

Sign in / Sign up

Export Citation Format

Share Document