Learning from Non-Seizure Clusters for EEG Analytics

Javad Birjandtalab, Melvin James, Mehrdad Nourani, Jay H Harvey

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

Abstract

EEG data collected in EMU is highly imbalanced and accuracy of automatic epileptic seizure detection is naturally low. Our aim is to increase the accuracy by reducing the imbalance ratio of seizure and non-seizure classes. We hypothesis that the non-seizure class itself includes various daily brain activities and then the data points are distributed as clusters in this class. In training phase, we propose a technique to cluster the majority (non-seizure) class into k clusters. Then, we train k KNN classifiers using each of k non-seizure clusters plus seizure class. In testing phase, we classify an incoming sample using this model and the non-seizure cluster closest to the incoming sample. We employed a state-of-the-art visualization technique to illustrate clusters of majority non-seizure class in two dimensions. The results, applied to MIT EEG dataset, show that our technique provides a higher average F-Measure accuracy.

Original languageEnglish (US)
Title of host publication2018 IEEE Biomedical Circuits and Systems Conference, BioCAS 2018 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538636039
DOIs
StatePublished - Dec 20 2018
Event2018 IEEE Biomedical Circuits and Systems Conference, BioCAS 2018 - Cleveland, United States
Duration: Oct 17 2018Oct 19 2018

Publication series

Name2018 IEEE Biomedical Circuits and Systems Conference, BioCAS 2018 - Proceedings

Conference

Conference2018 IEEE Biomedical Circuits and Systems Conference, BioCAS 2018
CountryUnited States
CityCleveland
Period10/17/1810/19/18

Fingerprint

electroencephalography
Electroencephalography
learning
Seizures
Learning
seizures
Epilepsy
Brain
Classifiers
Visualization
Testing
classifiers
brain
education
Datasets

Keywords

  • Clustering
  • Epilepsy
  • Imbalance Data
  • Majority Class
  • Spectral Analysis

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Health Informatics
  • Instrumentation
  • Signal Processing
  • Biomedical Engineering

Cite this

Birjandtalab, J., James, M., Nourani, M., & Harvey, J. H. (2018). Learning from Non-Seizure Clusters for EEG Analytics. In 2018 IEEE Biomedical Circuits and Systems Conference, BioCAS 2018 - Proceedings [8584837] (2018 IEEE Biomedical Circuits and Systems Conference, BioCAS 2018 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/BIOCAS.2018.8584837

Learning from Non-Seizure Clusters for EEG Analytics. / Birjandtalab, Javad; James, Melvin; Nourani, Mehrdad; Harvey, Jay H.

2018 IEEE Biomedical Circuits and Systems Conference, BioCAS 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2018. 8584837 (2018 IEEE Biomedical Circuits and Systems Conference, BioCAS 2018 - Proceedings).

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

Birjandtalab, J, James, M, Nourani, M & Harvey, JH 2018, Learning from Non-Seizure Clusters for EEG Analytics. in 2018 IEEE Biomedical Circuits and Systems Conference, BioCAS 2018 - Proceedings., 8584837, 2018 IEEE Biomedical Circuits and Systems Conference, BioCAS 2018 - Proceedings, Institute of Electrical and Electronics Engineers Inc., 2018 IEEE Biomedical Circuits and Systems Conference, BioCAS 2018, Cleveland, United States, 10/17/18. https://doi.org/10.1109/BIOCAS.2018.8584837
Birjandtalab J, James M, Nourani M, Harvey JH. Learning from Non-Seizure Clusters for EEG Analytics. In 2018 IEEE Biomedical Circuits and Systems Conference, BioCAS 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2018. 8584837. (2018 IEEE Biomedical Circuits and Systems Conference, BioCAS 2018 - Proceedings). https://doi.org/10.1109/BIOCAS.2018.8584837
Birjandtalab, Javad ; James, Melvin ; Nourani, Mehrdad ; Harvey, Jay H. / Learning from Non-Seizure Clusters for EEG Analytics. 2018 IEEE Biomedical Circuits and Systems Conference, BioCAS 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2018. (2018 IEEE Biomedical Circuits and Systems Conference, BioCAS 2018 - Proceedings).
@inproceedings{d97fe06516ab415b87d7f28cfb6206fb,
title = "Learning from Non-Seizure Clusters for EEG Analytics",
abstract = "EEG data collected in EMU is highly imbalanced and accuracy of automatic epileptic seizure detection is naturally low. Our aim is to increase the accuracy by reducing the imbalance ratio of seizure and non-seizure classes. We hypothesis that the non-seizure class itself includes various daily brain activities and then the data points are distributed as clusters in this class. In training phase, we propose a technique to cluster the majority (non-seizure) class into k clusters. Then, we train k KNN classifiers using each of k non-seizure clusters plus seizure class. In testing phase, we classify an incoming sample using this model and the non-seizure cluster closest to the incoming sample. We employed a state-of-the-art visualization technique to illustrate clusters of majority non-seizure class in two dimensions. The results, applied to MIT EEG dataset, show that our technique provides a higher average F-Measure accuracy.",
keywords = "Clustering, Epilepsy, Imbalance Data, Majority Class, Spectral Analysis",
author = "Javad Birjandtalab and Melvin James and Mehrdad Nourani and Harvey, {Jay H}",
year = "2018",
month = "12",
day = "20",
doi = "10.1109/BIOCAS.2018.8584837",
language = "English (US)",
series = "2018 IEEE Biomedical Circuits and Systems Conference, BioCAS 2018 - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
booktitle = "2018 IEEE Biomedical Circuits and Systems Conference, BioCAS 2018 - Proceedings",
address = "United States",

}

TY - GEN

T1 - Learning from Non-Seizure Clusters for EEG Analytics

AU - Birjandtalab, Javad

AU - James, Melvin

AU - Nourani, Mehrdad

AU - Harvey, Jay H

PY - 2018/12/20

Y1 - 2018/12/20

N2 - EEG data collected in EMU is highly imbalanced and accuracy of automatic epileptic seizure detection is naturally low. Our aim is to increase the accuracy by reducing the imbalance ratio of seizure and non-seizure classes. We hypothesis that the non-seizure class itself includes various daily brain activities and then the data points are distributed as clusters in this class. In training phase, we propose a technique to cluster the majority (non-seizure) class into k clusters. Then, we train k KNN classifiers using each of k non-seizure clusters plus seizure class. In testing phase, we classify an incoming sample using this model and the non-seizure cluster closest to the incoming sample. We employed a state-of-the-art visualization technique to illustrate clusters of majority non-seizure class in two dimensions. The results, applied to MIT EEG dataset, show that our technique provides a higher average F-Measure accuracy.

AB - EEG data collected in EMU is highly imbalanced and accuracy of automatic epileptic seizure detection is naturally low. Our aim is to increase the accuracy by reducing the imbalance ratio of seizure and non-seizure classes. We hypothesis that the non-seizure class itself includes various daily brain activities and then the data points are distributed as clusters in this class. In training phase, we propose a technique to cluster the majority (non-seizure) class into k clusters. Then, we train k KNN classifiers using each of k non-seizure clusters plus seizure class. In testing phase, we classify an incoming sample using this model and the non-seizure cluster closest to the incoming sample. We employed a state-of-the-art visualization technique to illustrate clusters of majority non-seizure class in two dimensions. The results, applied to MIT EEG dataset, show that our technique provides a higher average F-Measure accuracy.

KW - Clustering

KW - Epilepsy

KW - Imbalance Data

KW - Majority Class

KW - Spectral Analysis

UR - http://www.scopus.com/inward/record.url?scp=85060879775&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85060879775&partnerID=8YFLogxK

U2 - 10.1109/BIOCAS.2018.8584837

DO - 10.1109/BIOCAS.2018.8584837

M3 - Conference contribution

AN - SCOPUS:85060879775

T3 - 2018 IEEE Biomedical Circuits and Systems Conference, BioCAS 2018 - Proceedings

BT - 2018 IEEE Biomedical Circuits and Systems Conference, BioCAS 2018 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -