Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

A Self-Aware Epilepsy Monitoring System for Real-Time Epileptic Seizure Detection

Forooghifar, Farnaz ; Aminifar, Amir LU orcid ; Cammoun, Leila ; Wisniewski, Ilona ; Ciumas, Carolina ; Ryvlin, Philippe and Atienza, David (2022) In Mobile Networks and Applications 27(2). p.677-690
Abstract

Epilepsy is one of the most prevalent paroxystic neurological disorders that can dramatically degrade the quality of life and may even lead to death. Therefore, real-time epilepsy monitoring and seizure detection has become important over the past decades. In this context, wearable technologies offer a promising solution to pervasive epilepsy monitoring by removing the constraints with respect to time and location. In this paper, we propose a self-aware wearable system for real-time detection of epileptic seizures on a long-term basis. First, we propose a multi-parametric machine learning technique to detect seizures by analyzing both cardiac and respiratory responses to seizures, which are obtained using only the ECG signal. Second, in... (More)

Epilepsy is one of the most prevalent paroxystic neurological disorders that can dramatically degrade the quality of life and may even lead to death. Therefore, real-time epilepsy monitoring and seizure detection has become important over the past decades. In this context, wearable technologies offer a promising solution to pervasive epilepsy monitoring by removing the constraints with respect to time and location. In this paper, we propose a self-aware wearable system for real-time detection of epileptic seizures on a long-term basis. First, we propose a multi-parametric machine learning technique to detect seizures by analyzing both cardiac and respiratory responses to seizures, which are obtained using only the ECG signal. Second, in order to enable long-time epilepsy detection, we introduce the notion of self-awareness in our real-time wearable system. We evaluate the performance of our proposed solution based on an epilepsy database of more than 211 hours of recording, provided by the Lausanne University Hospital (CHUV), on the INYU wearable sensor. Our proposed system achieves a sensitivity of 88.66% and a specificity of 85.65% before applying self-awareness. Moreover, by controlling the energy-quality trade-offs using our self-aware energy-management technique, we can tune the battery lifetime of the wearable system to last between 67.55 and 136.91 days while, still outperforming the state-of-the-art techniques for wearable seizure detection, by achieving from 85.54% to 79.33% geometric mean of specificity and sensitivity.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; ; ; and
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Cardiac system, Electrocardiography, Epileptic seizure detection, Heart-rate variability, Respiratory system, Self-awareness
in
Mobile Networks and Applications
volume
27
issue
2
pages
677 - 690
publisher
Springer
external identifiers
  • scopus:85070338810
ISSN
1383-469X
DOI
10.1007/s11036-019-01322-7
language
English
LU publication?
no
additional info
Publisher Copyright: © 2019, Springer Science+Business Media, LLC, part of Springer Nature.
id
f8077d7b-a610-43bf-a101-99b500284ab5
date added to LUP
2022-02-05 01:23:12
date last changed
2022-06-30 10:41:20
@article{f8077d7b-a610-43bf-a101-99b500284ab5,
  abstract     = {{<p>Epilepsy is one of the most prevalent paroxystic neurological disorders that can dramatically degrade the quality of life and may even lead to death. Therefore, real-time epilepsy monitoring and seizure detection has become important over the past decades. In this context, wearable technologies offer a promising solution to pervasive epilepsy monitoring by removing the constraints with respect to time and location. In this paper, we propose a self-aware wearable system for real-time detection of epileptic seizures on a long-term basis. First, we propose a multi-parametric machine learning technique to detect seizures by analyzing both cardiac and respiratory responses to seizures, which are obtained using only the ECG signal. Second, in order to enable long-time epilepsy detection, we introduce the notion of self-awareness in our real-time wearable system. We evaluate the performance of our proposed solution based on an epilepsy database of more than 211 hours of recording, provided by the Lausanne University Hospital (CHUV), on the INYU wearable sensor. Our proposed system achieves a sensitivity of 88.66% and a specificity of 85.65% before applying self-awareness. Moreover, by controlling the energy-quality trade-offs using our self-aware energy-management technique, we can tune the battery lifetime of the wearable system to last between 67.55 and 136.91 days while, still outperforming the state-of-the-art techniques for wearable seizure detection, by achieving from 85.54% to 79.33% geometric mean of specificity and sensitivity.</p>}},
  author       = {{Forooghifar, Farnaz and Aminifar, Amir and Cammoun, Leila and Wisniewski, Ilona and Ciumas, Carolina and Ryvlin, Philippe and Atienza, David}},
  issn         = {{1383-469X}},
  keywords     = {{Cardiac system; Electrocardiography; Epileptic seizure detection; Heart-rate variability; Respiratory system; Self-awareness}},
  language     = {{eng}},
  number       = {{2}},
  pages        = {{677--690}},
  publisher    = {{Springer}},
  series       = {{Mobile Networks and Applications}},
  title        = {{A Self-Aware Epilepsy Monitoring System for Real-Time Epileptic Seizure Detection}},
  url          = {{http://dx.doi.org/10.1007/s11036-019-01322-7}},
  doi          = {{10.1007/s11036-019-01322-7}},
  volume       = {{27}},
  year         = {{2022}},
}