Epileptic seizure prediction using variational mixture of Gaussians

Ali Shahidi Zandi, Guy A. Dumont, Manouchehr Javidan, Reza Tafreshi

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

Abstract

We propose a novel patient-specific method for predicting epileptic seizures by analysis of positive zero-crossing intervals in scalp electroencephalogram (EEG). In real-time analysis, the histogram of these intervals for the current EEG epoch is computed, and the values which correspond to the bins discriminating between interictal and preictal references are selected as an observation. Then, the set of observations from the last 5 min is compared with two reference sets of data points (interictal and preictal) using a variational Gaussian mixture model (GMM) of the data, and a combined index is computed. Comparing this index with a patient-specific threshold, an alarm sequence is produced for each channel. Finally, a seizure prediction alarm is generated according to channel-based information. The proposed method was evaluated using ∼40.3 h of scalp EEG recordings from 6 patients with total of 28 partial seizures. A high sensitivity of 95% was achieved with a false prediction rate of 0.134/h and an average prediction time of 22.8 min for the test dataset.

Original languageEnglish
Title of host publication33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011
Pages7549-7552
Number of pages4
DOIs
Publication statusPublished - 2011
Event33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011 - Boston, MA, United States
Duration: 30 Aug 20113 Sep 2011

Other

Other33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011
CountryUnited States
CityBoston, MA
Period30/8/113/9/11

Fingerprint

Electroencephalography
Epilepsy
Scalp
Seizures
Bins
Observation
Datasets

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Signal Processing
  • Biomedical Engineering
  • Health Informatics

Cite this

Zandi, A. S., Dumont, G. A., Javidan, M., & Tafreshi, R. (2011). Epileptic seizure prediction using variational mixture of Gaussians. In 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011 (pp. 7549-7552). [6091861] https://doi.org/10.1109/IEMBS.2011.6091861

Epileptic seizure prediction using variational mixture of Gaussians. / Zandi, Ali Shahidi; Dumont, Guy A.; Javidan, Manouchehr; Tafreshi, Reza.

33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011. 2011. p. 7549-7552 6091861.

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

Zandi, AS, Dumont, GA, Javidan, M & Tafreshi, R 2011, Epileptic seizure prediction using variational mixture of Gaussians. in 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011., 6091861, pp. 7549-7552, 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011, Boston, MA, United States, 30/8/11. https://doi.org/10.1109/IEMBS.2011.6091861
Zandi AS, Dumont GA, Javidan M, Tafreshi R. Epileptic seizure prediction using variational mixture of Gaussians. In 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011. 2011. p. 7549-7552. 6091861 https://doi.org/10.1109/IEMBS.2011.6091861
Zandi, Ali Shahidi ; Dumont, Guy A. ; Javidan, Manouchehr ; Tafreshi, Reza. / Epileptic seizure prediction using variational mixture of Gaussians. 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011. 2011. pp. 7549-7552
@inproceedings{70f26180a85a478eb5ac9a58471dff3b,
title = "Epileptic seizure prediction using variational mixture of Gaussians",
abstract = "We propose a novel patient-specific method for predicting epileptic seizures by analysis of positive zero-crossing intervals in scalp electroencephalogram (EEG). In real-time analysis, the histogram of these intervals for the current EEG epoch is computed, and the values which correspond to the bins discriminating between interictal and preictal references are selected as an observation. Then, the set of observations from the last 5 min is compared with two reference sets of data points (interictal and preictal) using a variational Gaussian mixture model (GMM) of the data, and a combined index is computed. Comparing this index with a patient-specific threshold, an alarm sequence is produced for each channel. Finally, a seizure prediction alarm is generated according to channel-based information. The proposed method was evaluated using ∼40.3 h of scalp EEG recordings from 6 patients with total of 28 partial seizures. A high sensitivity of 95{\%} was achieved with a false prediction rate of 0.134/h and an average prediction time of 22.8 min for the test dataset.",
author = "Zandi, {Ali Shahidi} and Dumont, {Guy A.} and Manouchehr Javidan and Reza Tafreshi",
year = "2011",
doi = "10.1109/IEMBS.2011.6091861",
language = "English",
isbn = "9781424441211",
pages = "7549--7552",
booktitle = "33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011",

}

TY - GEN

T1 - Epileptic seizure prediction using variational mixture of Gaussians

AU - Zandi, Ali Shahidi

AU - Dumont, Guy A.

AU - Javidan, Manouchehr

AU - Tafreshi, Reza

PY - 2011

Y1 - 2011

N2 - We propose a novel patient-specific method for predicting epileptic seizures by analysis of positive zero-crossing intervals in scalp electroencephalogram (EEG). In real-time analysis, the histogram of these intervals for the current EEG epoch is computed, and the values which correspond to the bins discriminating between interictal and preictal references are selected as an observation. Then, the set of observations from the last 5 min is compared with two reference sets of data points (interictal and preictal) using a variational Gaussian mixture model (GMM) of the data, and a combined index is computed. Comparing this index with a patient-specific threshold, an alarm sequence is produced for each channel. Finally, a seizure prediction alarm is generated according to channel-based information. The proposed method was evaluated using ∼40.3 h of scalp EEG recordings from 6 patients with total of 28 partial seizures. A high sensitivity of 95% was achieved with a false prediction rate of 0.134/h and an average prediction time of 22.8 min for the test dataset.

AB - We propose a novel patient-specific method for predicting epileptic seizures by analysis of positive zero-crossing intervals in scalp electroencephalogram (EEG). In real-time analysis, the histogram of these intervals for the current EEG epoch is computed, and the values which correspond to the bins discriminating between interictal and preictal references are selected as an observation. Then, the set of observations from the last 5 min is compared with two reference sets of data points (interictal and preictal) using a variational Gaussian mixture model (GMM) of the data, and a combined index is computed. Comparing this index with a patient-specific threshold, an alarm sequence is produced for each channel. Finally, a seizure prediction alarm is generated according to channel-based information. The proposed method was evaluated using ∼40.3 h of scalp EEG recordings from 6 patients with total of 28 partial seizures. A high sensitivity of 95% was achieved with a false prediction rate of 0.134/h and an average prediction time of 22.8 min for the test dataset.

UR - http://www.scopus.com/inward/record.url?scp=84055213655&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84055213655&partnerID=8YFLogxK

U2 - 10.1109/IEMBS.2011.6091861

DO - 10.1109/IEMBS.2011.6091861

M3 - Conference contribution

SN - 9781424441211

SP - 7549

EP - 7552

BT - 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011

ER -