Towards automatic identification of epileptic recordings in long-term EEG monitoring
Author(s)
Kok, Xuen Hoong
Imtiaz, Syed Anas
Rodriguez Villegas, esther
Type
Conference Paper
Abstract
Electroencephalogram (EEG) is a crucial tool inthe diagnosis and management of epilepsy. The process ofanalyzing EEG is time consuming leading to the developmentof seizure detection algorithms to aid its analysis. This ap-proach is limited since it requires seizures to occur duringmonitoring periods and can often lead to misdiagnosis in caseswhere seizure occurrence is rare. For such cases, it has beenshown that the interictal periods in EEG signals, which is thepredominant state in long-term monitoring, can be useful forthe diagnosis of epilepsy. This paper presents an algorithm,using the information in interictal periods, to discriminatebetween long-term EEG recordings of epilepsy patients andhealthy subjects. It extracts several time and frequency-timedomain features from the signals and classifies them usingan ensemble classifier, achieving 100% sensitivity and 98.7%specificity in classifying 267 recordings from 105 subjects. Theresults demonstrate the feasibility of this approach to reliablyidentify EEG recordings of epilepsy subjects automaticallywhich can be highly useful to facilitate screening and diagnosisof epilepsy, especially in those parts of the world where thereis a lack of trained personnel for interpreting EEG signals.
Date Issued
2021-12-09
Date Acceptance
2021-07-15
Citation
2021, pp.273-276
Publisher
IEEE
Start Page
273
End Page
276
Copyright Statement
© 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Identifier
https://ieeexplore.ieee.org/document/9630782
Source
43rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society
Subjects
Science & Technology
Technology
Engineering, Biomedical
Engineering, Electrical & Electronic
Engineering
MISDIAGNOSIS
Algorithms
Electroencephalography
Epilepsy
Humans
Seizures
Humans
Epilepsy
Seizures
Electroencephalography
Algorithms
Publication Status
Published
Start Date
2021-10-31
Finish Date
2021-11-04
Coverage Spatial
Virtual
Date Publish Online
2021-12-09