Epileptic Seizure Detection: A Deep Learning Approach

Hussein, Ramy; Palangi, Hamid; Ward, Rabab K.; Wang, Z. Jane

doi:10.48550/arxiv.1803.09848

Cited by 12 publications

(32 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the cross-patient experiments, the average sensitivity, specificity and precision of 83.72%, 84.06% and 85.36% are respectively achieved, and the standard deviations being 0.1349, 0.1379 and 0.1020, respectively. These results exceed the current state-of-the-art performances on the noisy data of CHB-MIT in [17], [18] and [4]. The extensive experimental results show that the performance of the proposed new approach is promising and has high stability, with smaller variations compared to existing methods.…”

Section: Introductioncontrasting

confidence: 54%

“…Besides, though EEG signals are in general dynamic and non-linear, during a sufficiently small time period, the signal may be considered to be stationary. Based on the above three observations and inspired by an architecture in [18], we design a new approach by using bidirectional long shortterm memory (BiLSTM) integrated with an attention mechanism. Firstly, we introduce an attention mechanism over EEG channels.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Robust Deep Learning Approach for Automatic Classification of Seizures Against Non-seizures

Yao

et al. 2018

Preprint

View full text Add to dashboard Cite

Identifying epileptic seizures through analysis of the electroencephalography (EEG) signal becomes a standard method for the diagnosis of epilepsy. Manual seizure identification on EEG by trained neurologists is time-consuming, labor-intensive and error-prone, and a reliable automatic seizure/non-seizure classification method is needed. One of the challenges in automatic seizure/non-seizure classification is that seizure morphologies exhibit considerable variabilities. In order to capture essential seizure patterns, this paper leverages an attention mechanism and a bidirectional long short-term memory (BiLSTM) to exploit both spatial and temporal discriminating features and overcome seizure variabilities. The attention mechanism is to capture spatial features according to the contributions of different brain regions to seizures. The BiLSTM is to extract discriminating temporal features in the forward and the backward directions. Cross-validation experiments and cross-patient experiments over the noisy data of CHB-MIT are performed to evaluate our proposed approach. The obtained average sensitivity of 87.00%, specificity of 88.60% and precision of 88.63% in cross-validation experiments are higher than using the current state-of-the-art methods, and the standard deviations of our approach are lower. The evaluation results of cross-patient experiments indicate that, our approach has better performance compared with the current state-of-the-art methods and is more robust across patients.

show abstract

Section: Introductioncontrasting

confidence: 54%

Section: Introductionmentioning

confidence: 99%

A Robust Deep Learning Approach for Automatic Classification of Seizures Against Non-seizures

Yao

et al. 2018

Preprint

View full text Add to dashboard Cite

show abstract

“…Because seizure detection, which is often of a real-time flavor, is often treated as the seizure/non-seizure classification problem, many machine learning methods have been developed 7,[9][10][11][12][18][19][20][21] . Recently, deep learning techniques have been applied to the seizure detection problem 4,[13][14][15]22 . The evaluations of these methods are conducted with patientspecific or across-patients experiments.…”

Section: Related Workmentioning

confidence: 99%

“…Hussein et al design a deep neural network for seizure/non-seizure classification by using LSTM 22 . It extracts temporal features by using LSTM.…”

Section: Vidyaratne Et Al Propose a Deep Recurrent Architecture By Co...mentioning

confidence: 99%

See 1 more Smart Citation

A Novel Independent RNN Approach to Classification of Seizures against Non-seizures

Yao¹,

Cheng²,

Zhang³

2019

Preprint

View full text Add to dashboard Cite

In current clinical practices, electroencephalograms (EEG) are reviewed and analyzed by trained neurologists to provide supports for therapeutic decisions. Manual reviews can be laborious and error prone. Automatic and accurate seizure/non-seizure classification methods are desirable. A critical challenge is that seizure morphologies exhibit considerable variabilities. In order to capture essential seizure features, this paper leverages an emerging deep learning model, the independently recurrent neural network (IndRNN), to construct a new approach for the seizure/non-seizure classification. This new approach gradually expands the time scales, thereby extracting temporal and spatial features from the local time duration to the entire record. Evaluations are conducted with cross-validation experiments across subjects over the noisy data of CHB-MIT. Experimental results demonstrate that the proposed approach outperforms the current state-of-the-art methods. In addition, we explore how the segment length affects the classification performance. Thirteen different segment lengths are assessed, showing that the classification performance varies over the segment lengths, and the maximal fluctuating margin is more than 4%. Thus, the segment length is an important factor influencing the classification performance.

show abstract

The evaluation of the training, “the approach to epilepsy and epileptic seizure,” which was given to teachers in Türkiye

Kartal,

Karakaş,

Tuz

et al. 2024

Brain and Behavior

View full text Add to dashboard Cite

IntroductionEpilepsy is the most common neurological disorder among humans after headaches. According to the World Health Organization, approximately 50–65 million individuals were diagnosed with epilepsy throughout the world, and around two million new cases of epilepsy are added to this figure every year.MethodsDesigned as descriptive and cross‐sectional research, this study was performed on 132 elementary school teachers. Training on epilepsy and epileptic seizure was given to teachers. The pretest and posttest research data were collected with the face‐to‐face interview method. In this process, the epilepsy knowledge scale was used as well as a survey form that had questions designed to find out about teachers’ personal characteristics. The Statistical Package for Social Science 25.0 was utilized in the statistical analysis of research data. In the research, the statistical significance was identified if the p‐value was below.05 (p < .05).ResultsOf all teachers participating in the study, 59.1% were female, 90.2% were married, and 47.7% witnessed an epilepsy seizure before. The mean of teachers’ pretest epilepsy knowledge scores was 8.43 ± 4.31 points before the training while the mean of their posttest epilepsy knowledge scores was 12.65 ± 2.48 points after the training. The difference between the means of pretest and posttest scores was statistically significant (p = .000). After the training, there was a statistically significant increase in means of scores obtained by teachers from each item of the epilepsy knowledge scale (p < .05).ConclusionsAs there was a statistically significant improvement in levels of teachers’ knowledge about both epilepsy and epileptic seizure after the training, it is recommended that the training about the approach to epilepsy and epileptic seizure be given to all teachers, and additionally, including these topics in the course curricula of universities is recommended.

show abstract

Epileptic Seizure Detection: A Deep Learning Approach

Cited by 12 publications

References 55 publications

A Robust Deep Learning Approach for Automatic Classification of Seizures Against Non-seizures

A Robust Deep Learning Approach for Automatic Classification of Seizures Against Non-seizures

A Novel Independent RNN Approach to Classification of Seizures against Non-seizures

The evaluation of the training, “the approach to epilepsy and epileptic seizure,” which was given to teachers in Türkiye

Contact Info

Product

Resources

About