Epileptic seizure detection

K. Prabhakar Nayak, U. C. Niranjan

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Seizures are the phenomenon of rhythmic discharge from either a local area or the whole brain and the individual behavior usually lasts from seconds to minutes. Since seizures, in general, occur infrequently and unpredictably, automatic detection of seizures during long-term electroencephalograph (EEG) recordings is highly recommended. As EEG signals are non-stationary, the conventional methods of frequency analysis are not successful for diagnostic purposes. This paper presents a method of analysis of EEG signals, which is based on time-frequency analysis. Selected segments of the EEG signals are analyzed using time-frequency methods and the features are extracted for each segment. These features are used as an input to the artificial neural network (ANN), which provides the final classification of the EEG segments concerning the existence of seizures or not. Also, this paper presents another method of analysis of EEG signals based on K-means nearest neighbor classifier and the performance of this classifier is tested on a prior labeled EEG database consisting of normal and epileptic samples. The performance indicates overall accuracy from 84% to 98%.

Original languageEnglish
Title of host publication4th Kuala Lumpur International Conference on Biomedical Engineering 2008, Biomed 2008
Pages191-194
Number of pages4
Volume21 IFMBE
Edition1
DOIs
Publication statusPublished - 2008
Event4th Kuala Lumpur International Conference on Biomedical Engineering 2008, Biomed 2008 - Kuala Lumpur, Malaysia
Duration: 25-06-200828-06-2008

Conference

Conference4th Kuala Lumpur International Conference on Biomedical Engineering 2008, Biomed 2008
CountryMalaysia
CityKuala Lumpur
Period25-06-0828-06-08

Fingerprint

Classifiers
Brain
Neural networks

All Science Journal Classification (ASJC) codes

  • Biomedical Engineering
  • Bioengineering

Cite this

Nayak, K. P., & Niranjan, U. C. (2008). Epileptic seizure detection. In 4th Kuala Lumpur International Conference on Biomedical Engineering 2008, Biomed 2008 (1 ed., Vol. 21 IFMBE, pp. 191-194) https://doi.org/10.1007/978-3-540-69139-6-51
Nayak, K. Prabhakar ; Niranjan, U. C. / Epileptic seizure detection. 4th Kuala Lumpur International Conference on Biomedical Engineering 2008, Biomed 2008. Vol. 21 IFMBE 1. ed. 2008. pp. 191-194
@inproceedings{55949bae0bde4d67a314d8bd964178c0,
title = "Epileptic seizure detection",
abstract = "Seizures are the phenomenon of rhythmic discharge from either a local area or the whole brain and the individual behavior usually lasts from seconds to minutes. Since seizures, in general, occur infrequently and unpredictably, automatic detection of seizures during long-term electroencephalograph (EEG) recordings is highly recommended. As EEG signals are non-stationary, the conventional methods of frequency analysis are not successful for diagnostic purposes. This paper presents a method of analysis of EEG signals, which is based on time-frequency analysis. Selected segments of the EEG signals are analyzed using time-frequency methods and the features are extracted for each segment. These features are used as an input to the artificial neural network (ANN), which provides the final classification of the EEG segments concerning the existence of seizures or not. Also, this paper presents another method of analysis of EEG signals based on K-means nearest neighbor classifier and the performance of this classifier is tested on a prior labeled EEG database consisting of normal and epileptic samples. The performance indicates overall accuracy from 84{\%} to 98{\%}.",
author = "Nayak, {K. Prabhakar} and Niranjan, {U. C.}",
year = "2008",
doi = "10.1007/978-3-540-69139-6-51",
language = "English",
isbn = "9783540691389",
volume = "21 IFMBE",
pages = "191--194",
booktitle = "4th Kuala Lumpur International Conference on Biomedical Engineering 2008, Biomed 2008",
edition = "1",

}

Nayak, KP & Niranjan, UC 2008, Epileptic seizure detection. in 4th Kuala Lumpur International Conference on Biomedical Engineering 2008, Biomed 2008. 1 edn, vol. 21 IFMBE, pp. 191-194, 4th Kuala Lumpur International Conference on Biomedical Engineering 2008, Biomed 2008, Kuala Lumpur, Malaysia, 25-06-08. https://doi.org/10.1007/978-3-540-69139-6-51

Epileptic seizure detection. / Nayak, K. Prabhakar; Niranjan, U. C.

4th Kuala Lumpur International Conference on Biomedical Engineering 2008, Biomed 2008. Vol. 21 IFMBE 1. ed. 2008. p. 191-194.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Epileptic seizure detection

AU - Nayak, K. Prabhakar

AU - Niranjan, U. C.

PY - 2008

Y1 - 2008

N2 - Seizures are the phenomenon of rhythmic discharge from either a local area or the whole brain and the individual behavior usually lasts from seconds to minutes. Since seizures, in general, occur infrequently and unpredictably, automatic detection of seizures during long-term electroencephalograph (EEG) recordings is highly recommended. As EEG signals are non-stationary, the conventional methods of frequency analysis are not successful for diagnostic purposes. This paper presents a method of analysis of EEG signals, which is based on time-frequency analysis. Selected segments of the EEG signals are analyzed using time-frequency methods and the features are extracted for each segment. These features are used as an input to the artificial neural network (ANN), which provides the final classification of the EEG segments concerning the existence of seizures or not. Also, this paper presents another method of analysis of EEG signals based on K-means nearest neighbor classifier and the performance of this classifier is tested on a prior labeled EEG database consisting of normal and epileptic samples. The performance indicates overall accuracy from 84% to 98%.

AB - Seizures are the phenomenon of rhythmic discharge from either a local area or the whole brain and the individual behavior usually lasts from seconds to minutes. Since seizures, in general, occur infrequently and unpredictably, automatic detection of seizures during long-term electroencephalograph (EEG) recordings is highly recommended. As EEG signals are non-stationary, the conventional methods of frequency analysis are not successful for diagnostic purposes. This paper presents a method of analysis of EEG signals, which is based on time-frequency analysis. Selected segments of the EEG signals are analyzed using time-frequency methods and the features are extracted for each segment. These features are used as an input to the artificial neural network (ANN), which provides the final classification of the EEG segments concerning the existence of seizures or not. Also, this paper presents another method of analysis of EEG signals based on K-means nearest neighbor classifier and the performance of this classifier is tested on a prior labeled EEG database consisting of normal and epileptic samples. The performance indicates overall accuracy from 84% to 98%.

UR - http://www.scopus.com/inward/record.url?scp=78349262599&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78349262599&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-69139-6-51

DO - 10.1007/978-3-540-69139-6-51

M3 - Conference contribution

SN - 9783540691389

VL - 21 IFMBE

SP - 191

EP - 194

BT - 4th Kuala Lumpur International Conference on Biomedical Engineering 2008, Biomed 2008

ER -

Nayak KP, Niranjan UC. Epileptic seizure detection. In 4th Kuala Lumpur International Conference on Biomedical Engineering 2008, Biomed 2008. 1 ed. Vol. 21 IFMBE. 2008. p. 191-194 https://doi.org/10.1007/978-3-540-69139-6-51