Detection of neonatal seizure using multiple filters

M. S. Khlif, M. Mesbah, B. Boashashl, P. Colditz

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

2 Citations (Scopus)

Abstract

It is often impossible to accurately differentiate between seizure and non-seizure related activities in irifants based on clinical manifestations alone. The electroencephalogram (EEG) is therefore the best tool available for the recognition, management, and prognosis of neonatal seizures. The EEG signal is known to change structural characteristics between seizure and non-seizure states. In this work, matching pursuit (MP) decomposition, based on a coherent time-frequency (TF) dictionary, has provided us with a measure for quantifying changes in the structure of the neonatal EEG signal as it alternates between the various states. The quantification of state changes served as the basis for detecting seizures in 35 newborn patients. For each record, a patient-dependent threshold that marks the transition to seizure state is established. The use of multiple filters reduced the amount of artifacts and enhanced the detector performance. Overall, 93.4% detection accuracy and 0.26 false alarms per hour were achieved.

Original languageEnglish
Title of host publication10th International Conference on Information Sciences, Signal Processing and their Applications, ISSPA 2010
Pages284-287
Number of pages4
DOIs
Publication statusPublished - 2010
Event10th International Conference on Information Sciences, Signal Processing and their Applications, ISSPA 2010 - Kuala Lumpur, Malaysia
Duration: May 10 2010May 13 2010

Other

Other10th International Conference on Information Sciences, Signal Processing and their Applications, ISSPA 2010
CountryMalaysia
CityKuala Lumpur
Period5/10/105/13/10

Fingerprint

Electroencephalography
Glossaries
Detectors
Decomposition

Keywords

  • EEG
  • Matching pursuit
  • Newborn
  • Seizure
  • Time-frequency

ASJC Scopus subject areas

  • Computer Science Applications
  • Information Systems
  • Signal Processing

Cite this

Khlif, M. S., Mesbah, M., Boashashl, B., & Colditz, P. (2010). Detection of neonatal seizure using multiple filters. In 10th International Conference on Information Sciences, Signal Processing and their Applications, ISSPA 2010 (pp. 284-287). [5605469] https://doi.org/10.1109/ISSPA.2010.5605469

Detection of neonatal seizure using multiple filters. / Khlif, M. S.; Mesbah, M.; Boashashl, B.; Colditz, P.

10th International Conference on Information Sciences, Signal Processing and their Applications, ISSPA 2010. 2010. p. 284-287 5605469.

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

Khlif, MS, Mesbah, M, Boashashl, B & Colditz, P 2010, Detection of neonatal seizure using multiple filters. in 10th International Conference on Information Sciences, Signal Processing and their Applications, ISSPA 2010., 5605469, pp. 284-287, 10th International Conference on Information Sciences, Signal Processing and their Applications, ISSPA 2010, Kuala Lumpur, Malaysia, 5/10/10. https://doi.org/10.1109/ISSPA.2010.5605469
Khlif MS, Mesbah M, Boashashl B, Colditz P. Detection of neonatal seizure using multiple filters. In 10th International Conference on Information Sciences, Signal Processing and their Applications, ISSPA 2010. 2010. p. 284-287. 5605469 https://doi.org/10.1109/ISSPA.2010.5605469
Khlif, M. S. ; Mesbah, M. ; Boashashl, B. ; Colditz, P. / Detection of neonatal seizure using multiple filters. 10th International Conference on Information Sciences, Signal Processing and their Applications, ISSPA 2010. 2010. pp. 284-287
@inproceedings{8f9adc9c5d7644c0b2d073fb3cd6c6e6,
title = "Detection of neonatal seizure using multiple filters",
abstract = "It is often impossible to accurately differentiate between seizure and non-seizure related activities in irifants based on clinical manifestations alone. The electroencephalogram (EEG) is therefore the best tool available for the recognition, management, and prognosis of neonatal seizures. The EEG signal is known to change structural characteristics between seizure and non-seizure states. In this work, matching pursuit (MP) decomposition, based on a coherent time-frequency (TF) dictionary, has provided us with a measure for quantifying changes in the structure of the neonatal EEG signal as it alternates between the various states. The quantification of state changes served as the basis for detecting seizures in 35 newborn patients. For each record, a patient-dependent threshold that marks the transition to seizure state is established. The use of multiple filters reduced the amount of artifacts and enhanced the detector performance. Overall, 93.4{\%} detection accuracy and 0.26 false alarms per hour were achieved.",
keywords = "EEG, Matching pursuit, Newborn, Seizure, Time-frequency",
author = "Khlif, {M. S.} and M. Mesbah and B. Boashashl and P. Colditz",
year = "2010",
doi = "10.1109/ISSPA.2010.5605469",
language = "English",
isbn = "9781424471676",
pages = "284--287",
booktitle = "10th International Conference on Information Sciences, Signal Processing and their Applications, ISSPA 2010",

}

TY - GEN

T1 - Detection of neonatal seizure using multiple filters

AU - Khlif, M. S.

AU - Mesbah, M.

AU - Boashashl, B.

AU - Colditz, P.

PY - 2010

Y1 - 2010

N2 - It is often impossible to accurately differentiate between seizure and non-seizure related activities in irifants based on clinical manifestations alone. The electroencephalogram (EEG) is therefore the best tool available for the recognition, management, and prognosis of neonatal seizures. The EEG signal is known to change structural characteristics between seizure and non-seizure states. In this work, matching pursuit (MP) decomposition, based on a coherent time-frequency (TF) dictionary, has provided us with a measure for quantifying changes in the structure of the neonatal EEG signal as it alternates between the various states. The quantification of state changes served as the basis for detecting seizures in 35 newborn patients. For each record, a patient-dependent threshold that marks the transition to seizure state is established. The use of multiple filters reduced the amount of artifacts and enhanced the detector performance. Overall, 93.4% detection accuracy and 0.26 false alarms per hour were achieved.

AB - It is often impossible to accurately differentiate between seizure and non-seizure related activities in irifants based on clinical manifestations alone. The electroencephalogram (EEG) is therefore the best tool available for the recognition, management, and prognosis of neonatal seizures. The EEG signal is known to change structural characteristics between seizure and non-seizure states. In this work, matching pursuit (MP) decomposition, based on a coherent time-frequency (TF) dictionary, has provided us with a measure for quantifying changes in the structure of the neonatal EEG signal as it alternates between the various states. The quantification of state changes served as the basis for detecting seizures in 35 newborn patients. For each record, a patient-dependent threshold that marks the transition to seizure state is established. The use of multiple filters reduced the amount of artifacts and enhanced the detector performance. Overall, 93.4% detection accuracy and 0.26 false alarms per hour were achieved.

KW - EEG

KW - Matching pursuit

KW - Newborn

KW - Seizure

KW - Time-frequency

UR - http://www.scopus.com/inward/record.url?scp=78650272910&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78650272910&partnerID=8YFLogxK

U2 - 10.1109/ISSPA.2010.5605469

DO - 10.1109/ISSPA.2010.5605469

M3 - Conference contribution

SN - 9781424471676

SP - 284

EP - 287

BT - 10th International Conference on Information Sciences, Signal Processing and their Applications, ISSPA 2010

ER -