Advanced search
3 files | 1.48 MB Add to list

Towards improved design and evaluation of epileptic seizure predictors

Author
Organization
Abstract
Abstract—Objective: Key issues in the epilepsy seizure prediction research are (1) the reproducibility of results (2) the inability to compare multiple approaches directly. To overcome these problems, the Seizure Prediction Challenge was organized on Kaggle.com. It aimed at establishing benchmarks on a dataset with predefined train, validation and test sets. Our main objective is to analyse the competition format, and to propose improvements, which would facilitate a better comparison of algorithms. The second objective is to present a novel deep learning approach to seizure prediction and compare it to other commonly used methods using patient centered metrics. Methods: We used the competition’s datasets to illustrate the effects of data contamination. Having better data partitions, we compared three types of models in terms of different objectives. Results: We found that correct selection of test samples is crucial when evaluating the performance of seizure forecasting models. Moreover, we showed that models, which achieve state-of-the-art performance with respect to commonly used AUC, sensitivity and specificity metrics, may not yet be suitable for practical usage because of low precision scores. Conclusion: Correlation between validation and test datasets used in the competition limited its scientific value. Significance: Our findings provide guidelines which allow for a more objective evaluation of seizure prediction models.
Keywords
Epilepsy, linear discriminant analysis, neural networks, support vector machines

Downloads

  • (...).pdf
    • full text
    • |
    • UGent only
    • |
    • PDF
    • |
    • 491.98 KB
  • TBME2700086.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 491.98 KB
  • (...).pdf
    • full text
    • |
    • UGent only
    • |
    • PDF
    • |
    • 491.98 KB

Citation

Please use this url to cite or link to this publication:

MLA
Korshunova, Iryna et al. “Towards Improved Design and Evaluation of Epileptic Seizure Predictors.” IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING 65.3 (2018): 502–510. Print.
APA
Korshunova, I., Kindermans, P.-J., Degrave, J., Verhoeven, T., Brinkmann, B., & Dambre, J. (2018). Towards improved design and evaluation of epileptic seizure predictors. IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, 65(3), 502–510.
Chicago author-date
Korshunova, Iryna, Pieter-Jan Kindermans, Jonas Degrave, Thibault Verhoeven, Benjamin Brinkmann, and Joni Dambre. 2018. “Towards Improved Design and Evaluation of Epileptic Seizure Predictors.” Ieee Transactions on Biomedical Engineering 65 (3): 502–510.
Chicago author-date (all authors)
Korshunova, Iryna, Pieter-Jan Kindermans, Jonas Degrave, Thibault Verhoeven, Benjamin Brinkmann, and Joni Dambre. 2018. “Towards Improved Design and Evaluation of Epileptic Seizure Predictors.” Ieee Transactions on Biomedical Engineering 65 (3): 502–510.
Vancouver
1.
Korshunova I, Kindermans P-J, Degrave J, Verhoeven T, Brinkmann B, Dambre J. Towards improved design and evaluation of epileptic seizure predictors. IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING. IEEE; 2018;65(3):502–10.
IEEE
[1]
I. Korshunova, P.-J. Kindermans, J. Degrave, T. Verhoeven, B. Brinkmann, and J. Dambre, “Towards improved design and evaluation of epileptic seizure predictors,” IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, vol. 65, no. 3, pp. 502–510, 2018.
@article{8519033,
  abstract     = {Abstract—Objective: Key issues in the epilepsy seizure prediction research are (1) the reproducibility of results (2) the
inability to compare multiple approaches directly. To overcome these problems, the Seizure Prediction Challenge was organized on Kaggle.com. It aimed at establishing benchmarks on a dataset with predefined train, validation and test sets. Our main objective is to analyse the competition format, and to propose improvements, which would facilitate a better comparison of algorithms. The second objective is to present a novel deep learning approach to seizure prediction and compare it to other commonly used methods using patient centered metrics. Methods: We used the competition’s datasets to illustrate the effects of data contamination. Having better data partitions, we compared three types of models in terms of different objectives. Results: We found that correct selection of test samples is crucial when evaluating
the performance of seizure forecasting models. Moreover, we showed that models, which achieve state-of-the-art performance with respect to commonly used AUC, sensitivity and specificity metrics, may not yet be suitable for practical usage because of low precision scores. Conclusion: Correlation between validation and test datasets used in the competition limited its scientific value. Significance: Our findings provide guidelines which allow for a more objective evaluation of seizure prediction models.},
  author       = {Korshunova, Iryna and Kindermans, Pieter-Jan and Degrave, Jonas and Verhoeven, Thibault and Brinkmann, Benjamin and Dambre, Joni},
  issn         = {0018-9294},
  journal      = {IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING},
  keywords     = {Epilepsy,linear discriminant analysis,neural networks,support vector machines},
  language     = {eng},
  number       = {3},
  pages        = {502--510},
  publisher    = {IEEE},
  title        = {Towards improved design and evaluation of epileptic seizure predictors},
  url          = {http://dx.doi.org/10.1109/TBME.2017.2700086},
  volume       = {65},
  year         = {2018},
}

Altmetric
View in Altmetric
Web of Science
Times cited: