Automated detection of convulsive seizures using a wearable accelerometer device

Shitanshu Kusmakar, Chandan K. Karmakar, Bernard Yan, Terence J. O'Brien, Ramanathan Muthuganapathy, Marimuthu Palaniswami

Research output: Contribution to journalArticleResearchpeer-review

Abstract

Epileptic seizure detection requires specialized approaches such as video/electroencephalography monitoring. However, these approaches are restricted mainly to hospital setting and requires video/EEG analysis by experts, which makes these approaches resource- and labor-intensive. In contrast, we aim to develop a wireless remote monitoring system based on a single wrist-worn accelerometer device, which is sensitive to multiple types of convulsive seizures and is capable of detecting seizures with short duration. Simple time domain features including a new set of Poincaré plot based features were extracted from the active movement events recorded using a wrist-worn accelerometer device. The best features were then selected using the area under the ROC curve analysis. Kernelized support vector data description was then used to classify nonseizure and seizure events. The proposed algorithm was evaluated on 5576 h of recordings from 79 patients and detected 40 (86.95%) of 46 convulsive seizures (generalized tonic-clonic (GTCS), psychogenic nonepileptic, and complex partial seizures) from 20 patients with a total of 270 false alarms (1.16/24 h). Furthermore, the algorithm showed a comparable performance (sensitivity 95.23% and false alarm rate 0.64/24 h) with respect to existing unimodal and multimodal methods for GTCS detection. The promising results shows the potential to build an ambulatory monitoring convulsive seizure detection system. A wearable accelerometer based seizure detection system would aid in continuous assessment of convulsive seizures in a timely and non-invasive manner.

Original languageEnglish
Article number8378051
Pages (from-to)421-432
Number of pages12
JournalIEEE Transactions on Biomedical Engineering
Volume66
Issue number2
DOIs
Publication statusPublished - 1 Feb 2019
Externally publishedYes

Keywords

  • Accelerometer
  • complex partial seizures (CPS)
  • generalized tonic-clonic seizures (GTCS)
  • psychogenic non-epileptic seizures (PNES)
  • support vector data description (SVDD)
  • wearable device

Cite this

Kusmakar, Shitanshu ; Karmakar, Chandan K. ; Yan, Bernard ; O'Brien, Terence J. ; Muthuganapathy, Ramanathan ; Palaniswami, Marimuthu. / Automated detection of convulsive seizures using a wearable accelerometer device. In: IEEE Transactions on Biomedical Engineering. 2019 ; Vol. 66, No. 2. pp. 421-432.
@article{ed93ef5883ad4c449ccaee809ffd04ec,
title = "Automated detection of convulsive seizures using a wearable accelerometer device",
abstract = "Epileptic seizure detection requires specialized approaches such as video/electroencephalography monitoring. However, these approaches are restricted mainly to hospital setting and requires video/EEG analysis by experts, which makes these approaches resource- and labor-intensive. In contrast, we aim to develop a wireless remote monitoring system based on a single wrist-worn accelerometer device, which is sensitive to multiple types of convulsive seizures and is capable of detecting seizures with short duration. Simple time domain features including a new set of Poincar{\'e} plot based features were extracted from the active movement events recorded using a wrist-worn accelerometer device. The best features were then selected using the area under the ROC curve analysis. Kernelized support vector data description was then used to classify nonseizure and seizure events. The proposed algorithm was evaluated on 5576 h of recordings from 79 patients and detected 40 (86.95{\%}) of 46 convulsive seizures (generalized tonic-clonic (GTCS), psychogenic nonepileptic, and complex partial seizures) from 20 patients with a total of 270 false alarms (1.16/24 h). Furthermore, the algorithm showed a comparable performance (sensitivity 95.23{\%} and false alarm rate 0.64/24 h) with respect to existing unimodal and multimodal methods for GTCS detection. The promising results shows the potential to build an ambulatory monitoring convulsive seizure detection system. A wearable accelerometer based seizure detection system would aid in continuous assessment of convulsive seizures in a timely and non-invasive manner.",
keywords = "Accelerometer, complex partial seizures (CPS), generalized tonic-clonic seizures (GTCS), psychogenic non-epileptic seizures (PNES), support vector data description (SVDD), wearable device",
author = "Shitanshu Kusmakar and Karmakar, {Chandan K.} and Bernard Yan and O'Brien, {Terence J.} and Ramanathan Muthuganapathy and Marimuthu Palaniswami",
year = "2019",
month = "2",
day = "1",
doi = "10.1109/TBME.2018.2845865",
language = "English",
volume = "66",
pages = "421--432",
journal = "IEEE Transactions on Biomedical Engineering",
issn = "0018-9294",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
number = "2",

}

Kusmakar, S, Karmakar, CK, Yan, B, O'Brien, TJ, Muthuganapathy, R & Palaniswami, M 2019, 'Automated detection of convulsive seizures using a wearable accelerometer device' IEEE Transactions on Biomedical Engineering, vol. 66, no. 2, 8378051, pp. 421-432. https://doi.org/10.1109/TBME.2018.2845865

Automated detection of convulsive seizures using a wearable accelerometer device. / Kusmakar, Shitanshu; Karmakar, Chandan K.; Yan, Bernard; O'Brien, Terence J.; Muthuganapathy, Ramanathan; Palaniswami, Marimuthu.

In: IEEE Transactions on Biomedical Engineering, Vol. 66, No. 2, 8378051, 01.02.2019, p. 421-432.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Automated detection of convulsive seizures using a wearable accelerometer device

AU - Kusmakar, Shitanshu

AU - Karmakar, Chandan K.

AU - Yan, Bernard

AU - O'Brien, Terence J.

AU - Muthuganapathy, Ramanathan

AU - Palaniswami, Marimuthu

PY - 2019/2/1

Y1 - 2019/2/1

N2 - Epileptic seizure detection requires specialized approaches such as video/electroencephalography monitoring. However, these approaches are restricted mainly to hospital setting and requires video/EEG analysis by experts, which makes these approaches resource- and labor-intensive. In contrast, we aim to develop a wireless remote monitoring system based on a single wrist-worn accelerometer device, which is sensitive to multiple types of convulsive seizures and is capable of detecting seizures with short duration. Simple time domain features including a new set of Poincaré plot based features were extracted from the active movement events recorded using a wrist-worn accelerometer device. The best features were then selected using the area under the ROC curve analysis. Kernelized support vector data description was then used to classify nonseizure and seizure events. The proposed algorithm was evaluated on 5576 h of recordings from 79 patients and detected 40 (86.95%) of 46 convulsive seizures (generalized tonic-clonic (GTCS), psychogenic nonepileptic, and complex partial seizures) from 20 patients with a total of 270 false alarms (1.16/24 h). Furthermore, the algorithm showed a comparable performance (sensitivity 95.23% and false alarm rate 0.64/24 h) with respect to existing unimodal and multimodal methods for GTCS detection. The promising results shows the potential to build an ambulatory monitoring convulsive seizure detection system. A wearable accelerometer based seizure detection system would aid in continuous assessment of convulsive seizures in a timely and non-invasive manner.

AB - Epileptic seizure detection requires specialized approaches such as video/electroencephalography monitoring. However, these approaches are restricted mainly to hospital setting and requires video/EEG analysis by experts, which makes these approaches resource- and labor-intensive. In contrast, we aim to develop a wireless remote monitoring system based on a single wrist-worn accelerometer device, which is sensitive to multiple types of convulsive seizures and is capable of detecting seizures with short duration. Simple time domain features including a new set of Poincaré plot based features were extracted from the active movement events recorded using a wrist-worn accelerometer device. The best features were then selected using the area under the ROC curve analysis. Kernelized support vector data description was then used to classify nonseizure and seizure events. The proposed algorithm was evaluated on 5576 h of recordings from 79 patients and detected 40 (86.95%) of 46 convulsive seizures (generalized tonic-clonic (GTCS), psychogenic nonepileptic, and complex partial seizures) from 20 patients with a total of 270 false alarms (1.16/24 h). Furthermore, the algorithm showed a comparable performance (sensitivity 95.23% and false alarm rate 0.64/24 h) with respect to existing unimodal and multimodal methods for GTCS detection. The promising results shows the potential to build an ambulatory monitoring convulsive seizure detection system. A wearable accelerometer based seizure detection system would aid in continuous assessment of convulsive seizures in a timely and non-invasive manner.

KW - Accelerometer

KW - complex partial seizures (CPS)

KW - generalized tonic-clonic seizures (GTCS)

KW - psychogenic non-epileptic seizures (PNES)

KW - support vector data description (SVDD)

KW - wearable device

UR - http://www.scopus.com/inward/record.url?scp=85059046704&partnerID=8YFLogxK

U2 - 10.1109/TBME.2018.2845865

DO - 10.1109/TBME.2018.2845865

M3 - Article

VL - 66

SP - 421

EP - 432

JO - IEEE Transactions on Biomedical Engineering

JF - IEEE Transactions on Biomedical Engineering

SN - 0018-9294

IS - 2

M1 - 8378051

ER -