Review: Measuring saliency of features using signal-to-noise ratios for detection of electrocardiographic changes in partial epileptic patients

Egon van den Broek

Research output: Contribution to journalBook/Film/Article reviewAcademic

Abstract

Electrocardiographic (ECG) signals can be used to analyze and even predict patients’ epileptic attacks. However, this requires real-time processing and ECG signal classification. Using probabilistic neural networks (PNNs), a small set of salient ECG features is sufficient for an accurate classification of epileptic seizures. Übeyli has published various closely related studies, which diminishes the novelty of this paper. It starts with a concise overview of the signal-to-noise ratio (SNR) saliency measure, discrete wavelet transform (DWT), and PNN. Subsequently, the results on a standard dataset are presented: 98.33 percent accuracy, using only two features extracted by a DWT. Regrettably, the paper contains some flaws. It handles a two-class problem (for example, an epileptic seizure or not) and no multi-class problem. Hence, why a PNN is chosen as a classifier is not clear. Moreover, an experimental study like this should compare its processing scheme with a few related schemes. Finally, the results section lacks content; except for two accuracy percentages, hardly any information is provided. Although this paper has its limitations, it does pose a clear statement: the success of automatic classification depends on feature selection. Feature selection is also crucial in bringing offline pattern recognition to (online) real-time adaptable systems--for example, real-time detection of epileptic attacks. This is a message that cannot be repeated enough, since it is still too often forgotten.
Original languageUndefined
Pages (from-to)CR136893
Number of pages1
JournalComputing reviews
Publication statusPublished - 2 Jun 2009

Keywords

  • EWI-18401
  • discrete wavelet transform (DWT)
  • electrocardiography (ECG)
  • signal-to-noise ratio (SNR)
  • Epilepsy
  • HMI-CI: Computational Intelligence
  • HMI-HF: Human Factors
  • Review
  • probabilistic neural networks

Cite this

@article{db0c43cafa564a989ae7f87468fe34cf,
title = "Review: Measuring saliency of features using signal-to-noise ratios for detection of electrocardiographic changes in partial epileptic patients",
abstract = "Electrocardiographic (ECG) signals can be used to analyze and even predict patients’ epileptic attacks. However, this requires real-time processing and ECG signal classification. Using probabilistic neural networks (PNNs), a small set of salient ECG features is sufficient for an accurate classification of epileptic seizures. {\"U}beyli has published various closely related studies, which diminishes the novelty of this paper. It starts with a concise overview of the signal-to-noise ratio (SNR) saliency measure, discrete wavelet transform (DWT), and PNN. Subsequently, the results on a standard dataset are presented: 98.33 percent accuracy, using only two features extracted by a DWT. Regrettably, the paper contains some flaws. It handles a two-class problem (for example, an epileptic seizure or not) and no multi-class problem. Hence, why a PNN is chosen as a classifier is not clear. Moreover, an experimental study like this should compare its processing scheme with a few related schemes. Finally, the results section lacks content; except for two accuracy percentages, hardly any information is provided. Although this paper has its limitations, it does pose a clear statement: the success of automatic classification depends on feature selection. Feature selection is also crucial in bringing offline pattern recognition to (online) real-time adaptable systems--for example, real-time detection of epileptic attacks. This is a message that cannot be repeated enough, since it is still too often forgotten.",
keywords = "EWI-18401, discrete wavelet transform (DWT), electrocardiography (ECG), signal-to-noise ratio (SNR), Epilepsy, HMI-CI: Computational Intelligence, HMI-HF: Human Factors, Review, probabilistic neural networks",
author = "{van den Broek}, Egon",
year = "2009",
month = "6",
day = "2",
language = "Undefined",
pages = "CR136893",
journal = "Computing reviews",
issn = "0010-4884",
publisher = "Association for Computing Machinery (ACM)",

}

Review: Measuring saliency of features using signal-to-noise ratios for detection of electrocardiographic changes in partial epileptic patients. / van den Broek, Egon.

In: Computing reviews, 02.06.2009, p. CR136893.

Research output: Contribution to journalBook/Film/Article reviewAcademic

TY - JOUR

T1 - Review: Measuring saliency of features using signal-to-noise ratios for detection of electrocardiographic changes in partial epileptic patients

AU - van den Broek, Egon

PY - 2009/6/2

Y1 - 2009/6/2

N2 - Electrocardiographic (ECG) signals can be used to analyze and even predict patients’ epileptic attacks. However, this requires real-time processing and ECG signal classification. Using probabilistic neural networks (PNNs), a small set of salient ECG features is sufficient for an accurate classification of epileptic seizures. Übeyli has published various closely related studies, which diminishes the novelty of this paper. It starts with a concise overview of the signal-to-noise ratio (SNR) saliency measure, discrete wavelet transform (DWT), and PNN. Subsequently, the results on a standard dataset are presented: 98.33 percent accuracy, using only two features extracted by a DWT. Regrettably, the paper contains some flaws. It handles a two-class problem (for example, an epileptic seizure or not) and no multi-class problem. Hence, why a PNN is chosen as a classifier is not clear. Moreover, an experimental study like this should compare its processing scheme with a few related schemes. Finally, the results section lacks content; except for two accuracy percentages, hardly any information is provided. Although this paper has its limitations, it does pose a clear statement: the success of automatic classification depends on feature selection. Feature selection is also crucial in bringing offline pattern recognition to (online) real-time adaptable systems--for example, real-time detection of epileptic attacks. This is a message that cannot be repeated enough, since it is still too often forgotten.

AB - Electrocardiographic (ECG) signals can be used to analyze and even predict patients’ epileptic attacks. However, this requires real-time processing and ECG signal classification. Using probabilistic neural networks (PNNs), a small set of salient ECG features is sufficient for an accurate classification of epileptic seizures. Übeyli has published various closely related studies, which diminishes the novelty of this paper. It starts with a concise overview of the signal-to-noise ratio (SNR) saliency measure, discrete wavelet transform (DWT), and PNN. Subsequently, the results on a standard dataset are presented: 98.33 percent accuracy, using only two features extracted by a DWT. Regrettably, the paper contains some flaws. It handles a two-class problem (for example, an epileptic seizure or not) and no multi-class problem. Hence, why a PNN is chosen as a classifier is not clear. Moreover, an experimental study like this should compare its processing scheme with a few related schemes. Finally, the results section lacks content; except for two accuracy percentages, hardly any information is provided. Although this paper has its limitations, it does pose a clear statement: the success of automatic classification depends on feature selection. Feature selection is also crucial in bringing offline pattern recognition to (online) real-time adaptable systems--for example, real-time detection of epileptic attacks. This is a message that cannot be repeated enough, since it is still too often forgotten.

KW - EWI-18401

KW - discrete wavelet transform (DWT)

KW - electrocardiography (ECG)

KW - signal-to-noise ratio (SNR)

KW - Epilepsy

KW - HMI-CI: Computational Intelligence

KW - HMI-HF: Human Factors

KW - Review

KW - probabilistic neural networks

M3 - Book/Film/Article review

SP - CR136893

JO - Computing reviews

JF - Computing reviews

SN - 0010-4884

ER -