2016 17th International Conference on Sciences and Techniques of Automatic Control and Computer Engineering (STA) 2016
DOI: 10.1109/sta.2016.7952088
|View full text |Cite
|
Sign up to set email alerts
|

Seizure detection with single-channel EEG using Extreme Learning Machine

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
10
0

Year Published

2017
2017
2024
2024

Publication Types

Select...
4
2
1

Relationship

0
7

Authors

Journals

citations
Cited by 19 publications
(14 citation statements)
references
References 10 publications
0
10
0
Order By: Relevance
“…where Ψ (t) signifies the wavelet. WT was carried out to extract temporal measures in which spectral and temporal measures where the temporal features like mean, normalized coefficient of variation (NCOV), STD, skewness, kurtosis, spectral characteristics, mean PSD, and peak PSD were extracted [49]. DWT is used for the characterization of a signal as an infinite set of wavelets on an orthonormal basis [86].…”
Section: Continuous Wavelet Transform (Cwt)mentioning
confidence: 99%
See 2 more Smart Citations
“…where Ψ (t) signifies the wavelet. WT was carried out to extract temporal measures in which spectral and temporal measures where the temporal features like mean, normalized coefficient of variation (NCOV), STD, skewness, kurtosis, spectral characteristics, mean PSD, and peak PSD were extracted [49]. DWT is used for the characterization of a signal as an infinite set of wavelets on an orthonormal basis [86].…”
Section: Continuous Wavelet Transform (Cwt)mentioning
confidence: 99%
“…Epileptic seizure detection was performed with wavelet-based feature and time domain features employing the SVM classifier [48]. Time-frequency domain feature extraction was done, and an ELM classifier was utilized to distinguish seizures from non-seizures [49]. Three features were extracted, and classification was provided by six well-known classifiers, namely RF classifier, Functional tree (FT) classifier, K-NN, C4.5 classifier, NB, and Bayes Net [50].…”
Section: Classificationmentioning
confidence: 99%
See 1 more Smart Citation
“…Traditionally, Digital filters are used to remove noise from the observed signal and to reduce its undesirable frequency components [3]. Single channel EEG signals were analysed for seizure detection in [6] and [7]. An accuracy of over 90% was achieved in the former after experimenting with multiple classification primitives and SVM ensemble parameters.…”
Section: Figmentioning
confidence: 99%
“…Extreme Learning Machines [7] and Support Vector Machine ensembles [6], which were used with single channel EEG signals can be applied to multi-channel EEG signals as a part of further research. Whether or not using the leaky ReLU activation function along with the BRNN can lead to further improvement in training stage is another area for further research.…”
Section: Potential For Further Researchmentioning
confidence: 99%