Automatic seizure detection in multichannel EEG using McCIT2FIS approach

S. Dora, B. Rangarajan, K. Subramanian, S. Suresh

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

Abstract

In this paper, an automatic seizure detection technique using multichannel EEG is proposed based on Metacognitive Complex-valued Interval Type-2 Fuzzy Inference System (McCIT2FIS). A wavelet chaos theory based feature extraction is employed to extract the features from EEG signal as it can handle the non stationarity in data and Sparse Multinomial Logistic Regression via Bayesian L1 Regularisation (SBMLR) based feature selection is employed to select the most discriminative features. McCIT2FIS is employed to classify the samples as either interictal or ictal EEG segment as it has been shown to be capable of handling noisy data by virtue of Interval Type-2 fuzzy sets, and is good at classification because of its ability to handle complex-valued data. Further, we have also shown that the feature selected using SBMLR can be successfully mapped back to the channels allowing us to identify the epileptogenic regions of the brain. The performance of the McCIT2FIS was also compared with the support vector machines and the results indicate that McCIT2FIS is better capable of detecting seizure based on EEG signals.

Original languageEnglish
Title of host publicationFUZZ-IEEE 2015 - IEEE International Conference on Fuzzy Systems
EditorsAdnan Yazici, Nikhil R. Pal, Hisao Ishibuchi, Bulent Tutmez, Chin-Teng Lin, Joao M. C. Sousa, Uzay Kaymak, Trevor Martin
PublisherInstitute of Electrical and Electronics Engineers Inc.
Volume2015-November
ISBN (Electronic)9781467374286
DOIs
Publication statusPublished - 25 Nov 2015
EventIEEE International Conference on Fuzzy Systems, FUZZ-IEEE 2015 - Istanbul, Turkey
Duration: 2 Aug 20155 Aug 2015

Conference

ConferenceIEEE International Conference on Fuzzy Systems, FUZZ-IEEE 2015
CountryTurkey
CityIstanbul
Period2/08/155/08/15

Fingerprint

Fuzzy Inference System
Fuzzy inference
Electroencephalography
Interval
Feature extraction
Data handling
Interval type-2 Fuzzy Sets
Fuzzy sets
Chaos Theory
Chaos theory
Nonstationarity
Support vector machines
Logistics
Noisy Data
Brain
Logistic Regression
Feature Selection
Feature Extraction
Support Vector Machine
Regularization

Keywords

  • complex-valued fuzzy system
  • electroencephalogram
  • interval type-2
  • noisy data
  • Seizure detection

Cite this

Dora, S., Rangarajan, B., Subramanian, K., & Suresh, S. (2015). Automatic seizure detection in multichannel EEG using McCIT2FIS approach. In A. Yazici, N. R. Pal, H. Ishibuchi, B. Tutmez, C-T. Lin, J. M. C. Sousa, U. Kaymak, ... T. Martin (Eds.), FUZZ-IEEE 2015 - IEEE International Conference on Fuzzy Systems (Vol. 2015-November). [7338085] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/FUZZ-IEEE.2015.7338085
Dora, S. ; Rangarajan, B. ; Subramanian, K. ; Suresh, S. / Automatic seizure detection in multichannel EEG using McCIT2FIS approach. FUZZ-IEEE 2015 - IEEE International Conference on Fuzzy Systems. editor / Adnan Yazici ; Nikhil R. Pal ; Hisao Ishibuchi ; Bulent Tutmez ; Chin-Teng Lin ; Joao M. C. Sousa ; Uzay Kaymak ; Trevor Martin. Vol. 2015-November Institute of Electrical and Electronics Engineers Inc., 2015.
@inproceedings{1a80d568d69344218a46b014be8a3c8b,
title = "Automatic seizure detection in multichannel EEG using McCIT2FIS approach",
abstract = "In this paper, an automatic seizure detection technique using multichannel EEG is proposed based on Metacognitive Complex-valued Interval Type-2 Fuzzy Inference System (McCIT2FIS). A wavelet chaos theory based feature extraction is employed to extract the features from EEG signal as it can handle the non stationarity in data and Sparse Multinomial Logistic Regression via Bayesian L1 Regularisation (SBMLR) based feature selection is employed to select the most discriminative features. McCIT2FIS is employed to classify the samples as either interictal or ictal EEG segment as it has been shown to be capable of handling noisy data by virtue of Interval Type-2 fuzzy sets, and is good at classification because of its ability to handle complex-valued data. Further, we have also shown that the feature selected using SBMLR can be successfully mapped back to the channels allowing us to identify the epileptogenic regions of the brain. The performance of the McCIT2FIS was also compared with the support vector machines and the results indicate that McCIT2FIS is better capable of detecting seizure based on EEG signals.",
keywords = "complex-valued fuzzy system, electroencephalogram, interval type-2, noisy data, Seizure detection",
author = "S. Dora and B. Rangarajan and K. Subramanian and S. Suresh",
year = "2015",
month = "11",
day = "25",
doi = "10.1109/FUZZ-IEEE.2015.7338085",
language = "English",
volume = "2015-November",
editor = "Adnan Yazici and Pal, {Nikhil R.} and Hisao Ishibuchi and Bulent Tutmez and Chin-Teng Lin and Sousa, {Joao M. C.} and Uzay Kaymak and Trevor Martin",
booktitle = "FUZZ-IEEE 2015 - IEEE International Conference on Fuzzy Systems",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

Dora, S, Rangarajan, B, Subramanian, K & Suresh, S 2015, Automatic seizure detection in multichannel EEG using McCIT2FIS approach. in A Yazici, NR Pal, H Ishibuchi, B Tutmez, C-T Lin, JMC Sousa, U Kaymak & T Martin (eds), FUZZ-IEEE 2015 - IEEE International Conference on Fuzzy Systems. vol. 2015-November, 7338085, Institute of Electrical and Electronics Engineers Inc., IEEE International Conference on Fuzzy Systems, FUZZ-IEEE 2015, Istanbul, Turkey, 2/08/15. https://doi.org/10.1109/FUZZ-IEEE.2015.7338085

Automatic seizure detection in multichannel EEG using McCIT2FIS approach. / Dora, S.; Rangarajan, B.; Subramanian, K.; Suresh, S.

FUZZ-IEEE 2015 - IEEE International Conference on Fuzzy Systems. ed. / Adnan Yazici; Nikhil R. Pal; Hisao Ishibuchi; Bulent Tutmez; Chin-Teng Lin; Joao M. C. Sousa; Uzay Kaymak; Trevor Martin. Vol. 2015-November Institute of Electrical and Electronics Engineers Inc., 2015. 7338085.

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

TY - GEN

T1 - Automatic seizure detection in multichannel EEG using McCIT2FIS approach

AU - Dora, S.

AU - Rangarajan, B.

AU - Subramanian, K.

AU - Suresh, S.

PY - 2015/11/25

Y1 - 2015/11/25

N2 - In this paper, an automatic seizure detection technique using multichannel EEG is proposed based on Metacognitive Complex-valued Interval Type-2 Fuzzy Inference System (McCIT2FIS). A wavelet chaos theory based feature extraction is employed to extract the features from EEG signal as it can handle the non stationarity in data and Sparse Multinomial Logistic Regression via Bayesian L1 Regularisation (SBMLR) based feature selection is employed to select the most discriminative features. McCIT2FIS is employed to classify the samples as either interictal or ictal EEG segment as it has been shown to be capable of handling noisy data by virtue of Interval Type-2 fuzzy sets, and is good at classification because of its ability to handle complex-valued data. Further, we have also shown that the feature selected using SBMLR can be successfully mapped back to the channels allowing us to identify the epileptogenic regions of the brain. The performance of the McCIT2FIS was also compared with the support vector machines and the results indicate that McCIT2FIS is better capable of detecting seizure based on EEG signals.

AB - In this paper, an automatic seizure detection technique using multichannel EEG is proposed based on Metacognitive Complex-valued Interval Type-2 Fuzzy Inference System (McCIT2FIS). A wavelet chaos theory based feature extraction is employed to extract the features from EEG signal as it can handle the non stationarity in data and Sparse Multinomial Logistic Regression via Bayesian L1 Regularisation (SBMLR) based feature selection is employed to select the most discriminative features. McCIT2FIS is employed to classify the samples as either interictal or ictal EEG segment as it has been shown to be capable of handling noisy data by virtue of Interval Type-2 fuzzy sets, and is good at classification because of its ability to handle complex-valued data. Further, we have also shown that the feature selected using SBMLR can be successfully mapped back to the channels allowing us to identify the epileptogenic regions of the brain. The performance of the McCIT2FIS was also compared with the support vector machines and the results indicate that McCIT2FIS is better capable of detecting seizure based on EEG signals.

KW - complex-valued fuzzy system

KW - electroencephalogram

KW - interval type-2

KW - noisy data

KW - Seizure detection

UR - http://www.scopus.com/inward/record.url?scp=84975770169&partnerID=8YFLogxK

U2 - 10.1109/FUZZ-IEEE.2015.7338085

DO - 10.1109/FUZZ-IEEE.2015.7338085

M3 - Conference contribution

AN - SCOPUS:84975770169

VL - 2015-November

BT - FUZZ-IEEE 2015 - IEEE International Conference on Fuzzy Systems

A2 - Yazici, Adnan

A2 - Pal, Nikhil R.

A2 - Ishibuchi, Hisao

A2 - Tutmez, Bulent

A2 - Lin, Chin-Teng

A2 - Sousa, Joao M. C.

A2 - Kaymak, Uzay

A2 - Martin, Trevor

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Dora S, Rangarajan B, Subramanian K, Suresh S. Automatic seizure detection in multichannel EEG using McCIT2FIS approach. In Yazici A, Pal NR, Ishibuchi H, Tutmez B, Lin C-T, Sousa JMC, Kaymak U, Martin T, editors, FUZZ-IEEE 2015 - IEEE International Conference on Fuzzy Systems. Vol. 2015-November. Institute of Electrical and Electronics Engineers Inc. 2015. 7338085 https://doi.org/10.1109/FUZZ-IEEE.2015.7338085