Adaptive parametric spectral estimation with Kalman smoothing for online early seizure detection

Yun S. Park, Leigh R. Hochberg, Emad N. Eskandar, Sydney S. Cash, Wilson Truccolo

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

4 Citations (Scopus)

Abstract

Tracking spectral changes in neural signals, such as local field potentials (LFPs) and scalp or intracranial electroencephalograms (EEG, iEEG), is an important problem in early detection and prediction of seizures. Most approaches have focused on either parametric or nonparametric spectral estimation methods based on moving time windows. Here, we explore an adaptive (time-varying) parametric ARMA approach for tracking spectral changes in neural signals based on the fixed-interval Kalman smoother. We apply the method to seizure detection based on spectral features of intracortical LFPs recorded from a person with pharmacologically intractable focal epilepsy. We also devise and test an approach for real-time tracking of spectra based on the adaptive parametric method with the fixed-interval Kalman smoother. The order of ARMA models is determined via the AIC computed in moving time windows. We quantitatively demonstrate the advantages of using the adaptive parametric estimation method in seizure detection over nonparametric alternatives based exclusively on moving time windows. Overall, the adaptive parametric approach significantly improves the statistical separability of interictal and ictal epochs.

Original languageEnglish (US)
Title of host publication2013 6th International IEEE EMBS Conference on Neural Engineering, NER 2013
Pages1410-1413
Number of pages4
DOIs
StatePublished - Dec 1 2013
Externally publishedYes
Event2013 6th International IEEE EMBS Conference on Neural Engineering, NER 2013 - San Diego, CA, United States
Duration: Nov 6 2013Nov 8 2013

Other

Other2013 6th International IEEE EMBS Conference on Neural Engineering, NER 2013
CountryUnited States
CitySan Diego, CA
Period11/6/1311/8/13

Fingerprint

Electroencephalography
Bioelectric potentials

ASJC Scopus subject areas

  • Artificial Intelligence
  • Mechanical Engineering

Cite this

Park, Y. S., Hochberg, L. R., Eskandar, E. N., Cash, S. S., & Truccolo, W. (2013). Adaptive parametric spectral estimation with Kalman smoothing for online early seizure detection. In 2013 6th International IEEE EMBS Conference on Neural Engineering, NER 2013 (pp. 1410-1413). [6696207] https://doi.org/10.1109/NER.2013.6696207

Adaptive parametric spectral estimation with Kalman smoothing for online early seizure detection. / Park, Yun S.; Hochberg, Leigh R.; Eskandar, Emad N.; Cash, Sydney S.; Truccolo, Wilson.

2013 6th International IEEE EMBS Conference on Neural Engineering, NER 2013. 2013. p. 1410-1413 6696207.

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

Park, YS, Hochberg, LR, Eskandar, EN, Cash, SS & Truccolo, W 2013, Adaptive parametric spectral estimation with Kalman smoothing for online early seizure detection. in 2013 6th International IEEE EMBS Conference on Neural Engineering, NER 2013., 6696207, pp. 1410-1413, 2013 6th International IEEE EMBS Conference on Neural Engineering, NER 2013, San Diego, CA, United States, 11/6/13. https://doi.org/10.1109/NER.2013.6696207
Park YS, Hochberg LR, Eskandar EN, Cash SS, Truccolo W. Adaptive parametric spectral estimation with Kalman smoothing for online early seizure detection. In 2013 6th International IEEE EMBS Conference on Neural Engineering, NER 2013. 2013. p. 1410-1413. 6696207 https://doi.org/10.1109/NER.2013.6696207
Park, Yun S. ; Hochberg, Leigh R. ; Eskandar, Emad N. ; Cash, Sydney S. ; Truccolo, Wilson. / Adaptive parametric spectral estimation with Kalman smoothing for online early seizure detection. 2013 6th International IEEE EMBS Conference on Neural Engineering, NER 2013. 2013. pp. 1410-1413
@inproceedings{f2fd58aabbce4f2589d3130faf789f66,
title = "Adaptive parametric spectral estimation with Kalman smoothing for online early seizure detection",
abstract = "Tracking spectral changes in neural signals, such as local field potentials (LFPs) and scalp or intracranial electroencephalograms (EEG, iEEG), is an important problem in early detection and prediction of seizures. Most approaches have focused on either parametric or nonparametric spectral estimation methods based on moving time windows. Here, we explore an adaptive (time-varying) parametric ARMA approach for tracking spectral changes in neural signals based on the fixed-interval Kalman smoother. We apply the method to seizure detection based on spectral features of intracortical LFPs recorded from a person with pharmacologically intractable focal epilepsy. We also devise and test an approach for real-time tracking of spectra based on the adaptive parametric method with the fixed-interval Kalman smoother. The order of ARMA models is determined via the AIC computed in moving time windows. We quantitatively demonstrate the advantages of using the adaptive parametric estimation method in seizure detection over nonparametric alternatives based exclusively on moving time windows. Overall, the adaptive parametric approach significantly improves the statistical separability of interictal and ictal epochs.",
author = "Park, {Yun S.} and Hochberg, {Leigh R.} and Eskandar, {Emad N.} and Cash, {Sydney S.} and Wilson Truccolo",
year = "2013",
month = "12",
day = "1",
doi = "10.1109/NER.2013.6696207",
language = "English (US)",
isbn = "9781467319690",
pages = "1410--1413",
booktitle = "2013 6th International IEEE EMBS Conference on Neural Engineering, NER 2013",

}

TY - GEN

T1 - Adaptive parametric spectral estimation with Kalman smoothing for online early seizure detection

AU - Park, Yun S.

AU - Hochberg, Leigh R.

AU - Eskandar, Emad N.

AU - Cash, Sydney S.

AU - Truccolo, Wilson

PY - 2013/12/1

Y1 - 2013/12/1

N2 - Tracking spectral changes in neural signals, such as local field potentials (LFPs) and scalp or intracranial electroencephalograms (EEG, iEEG), is an important problem in early detection and prediction of seizures. Most approaches have focused on either parametric or nonparametric spectral estimation methods based on moving time windows. Here, we explore an adaptive (time-varying) parametric ARMA approach for tracking spectral changes in neural signals based on the fixed-interval Kalman smoother. We apply the method to seizure detection based on spectral features of intracortical LFPs recorded from a person with pharmacologically intractable focal epilepsy. We also devise and test an approach for real-time tracking of spectra based on the adaptive parametric method with the fixed-interval Kalman smoother. The order of ARMA models is determined via the AIC computed in moving time windows. We quantitatively demonstrate the advantages of using the adaptive parametric estimation method in seizure detection over nonparametric alternatives based exclusively on moving time windows. Overall, the adaptive parametric approach significantly improves the statistical separability of interictal and ictal epochs.

AB - Tracking spectral changes in neural signals, such as local field potentials (LFPs) and scalp or intracranial electroencephalograms (EEG, iEEG), is an important problem in early detection and prediction of seizures. Most approaches have focused on either parametric or nonparametric spectral estimation methods based on moving time windows. Here, we explore an adaptive (time-varying) parametric ARMA approach for tracking spectral changes in neural signals based on the fixed-interval Kalman smoother. We apply the method to seizure detection based on spectral features of intracortical LFPs recorded from a person with pharmacologically intractable focal epilepsy. We also devise and test an approach for real-time tracking of spectra based on the adaptive parametric method with the fixed-interval Kalman smoother. The order of ARMA models is determined via the AIC computed in moving time windows. We quantitatively demonstrate the advantages of using the adaptive parametric estimation method in seizure detection over nonparametric alternatives based exclusively on moving time windows. Overall, the adaptive parametric approach significantly improves the statistical separability of interictal and ictal epochs.

UR - http://www.scopus.com/inward/record.url?scp=84897713831&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84897713831&partnerID=8YFLogxK

U2 - 10.1109/NER.2013.6696207

DO - 10.1109/NER.2013.6696207

M3 - Conference contribution

SN - 9781467319690

SP - 1410

EP - 1413

BT - 2013 6th International IEEE EMBS Conference on Neural Engineering, NER 2013

ER -