Morphology-based detection of premature ventricular contractions

Rohit Hadia, Daniel Guldenring, Dewar Finlay, Alan Kennedy, Ghalib Janjua, Raymond Bond, James McLaughlin

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Premature ventricular contraction (PVC) is the type of ectopic heartbeat, commonly found in the healthy population and is often considered benign. However, they are reported to adversely affect the accuracy of R-R variability based electrocardiographic (ECG) algorithms. This study proposes a Principal Component Analysis (PCA) based algorithmic approach to detect the PVCs based on their morphology. The eigenvectors were derived from signal window around the R-peak, where signal window for the PVC (wPVC) and that of NSR (wNSR) were set to 0.55 seconds and 0.16 seconds respectively. We used 24 ECG recordings from MIT BIH arrhythmia database as training dataset and the remaining 24 ECG recordings as testing dataset. Using the derived eigenvectors and the Linear regression (LR) analysis; complexes corresponding to the wNSR and wPVC were estimated from training and testing datasets. Four different classification methods were employed to differentiate between wPVS and wNSR, namely, Root mean squared error (RMSE), Pearson product-moment correlation coefficient comparision, Histogram probability distribution and k-Nearest Neighbour (KNN). All four methods were implemented individually to classify the wPVC and wNSR. The performance of each of the classification approach was evaluated by computing sensitivity and specificity. With the sensitivity of 93.45% and specificity of 93.14%, KNN based classification method has shown the best performance. The method proposed in this study allows for an effective differentiation between NSR beats and PVC beats.
LanguageEnglish
Title of host publicationUnknown Host Publication
Number of pages4
Publication statusAccepted/In press - 1 Jun 2017
EventComputing in Cardiology - Rennes, France
Duration: 1 Jun 2017 → …

Conference

ConferenceComputing in Cardiology
Period1/06/17 → …

Fingerprint

Polyvinyl chlorides
Eigenvalues and eigenfunctions
Testing
Linear regression
Regression analysis
Principal component analysis
Probability distributions

Keywords

  • Electrocardiography
  • Principal component analysis
  • Heart rate variability
  • Morphology
  • Training
  • Testing
  • Correlation coefficient

Cite this

Hadia, R., Guldenring, D., Finlay, D., Kennedy, A., Janjua, G., Bond, R., & McLaughlin, J. (Accepted/In press). Morphology-based detection of premature ventricular contractions. In Unknown Host Publication
Hadia, Rohit ; Guldenring, Daniel ; Finlay, Dewar ; Kennedy, Alan ; Janjua, Ghalib ; Bond, Raymond ; McLaughlin, James. / Morphology-based detection of premature ventricular contractions. Unknown Host Publication. 2017.
@inproceedings{8d336217db984c90b60f9b9619ada970,
title = "Morphology-based detection of premature ventricular contractions",
abstract = "Premature ventricular contraction (PVC) is the type of ectopic heartbeat, commonly found in the healthy population and is often considered benign. However, they are reported to adversely affect the accuracy of R-R variability based electrocardiographic (ECG) algorithms. This study proposes a Principal Component Analysis (PCA) based algorithmic approach to detect the PVCs based on their morphology. The eigenvectors were derived from signal window around the R-peak, where signal window for the PVC (wPVC) and that of NSR (wNSR) were set to 0.55 seconds and 0.16 seconds respectively. We used 24 ECG recordings from MIT BIH arrhythmia database as training dataset and the remaining 24 ECG recordings as testing dataset. Using the derived eigenvectors and the Linear regression (LR) analysis; complexes corresponding to the wNSR and wPVC were estimated from training and testing datasets. Four different classification methods were employed to differentiate between wPVS and wNSR, namely, Root mean squared error (RMSE), Pearson product-moment correlation coefficient comparision, Histogram probability distribution and k-Nearest Neighbour (KNN). All four methods were implemented individually to classify the wPVC and wNSR. The performance of each of the classification approach was evaluated by computing sensitivity and specificity. With the sensitivity of 93.45{\%} and specificity of 93.14{\%}, KNN based classification method has shown the best performance. The method proposed in this study allows for an effective differentiation between NSR beats and PVC beats.",
keywords = "Electrocardiography, Principal component analysis, Heart rate variability, Morphology, Training, Testing, Correlation coefficient",
author = "Rohit Hadia and Daniel Guldenring and Dewar Finlay and Alan Kennedy and Ghalib Janjua and Raymond Bond and James McLaughlin",
year = "2017",
month = "6",
day = "1",
language = "English",
isbn = "978-1-5386-6630-2",
booktitle = "Unknown Host Publication",

}

Hadia, R, Guldenring, D, Finlay, D, Kennedy, A, Janjua, G, Bond, R & McLaughlin, J 2017, Morphology-based detection of premature ventricular contractions. in Unknown Host Publication. Computing in Cardiology, 1/06/17.

Morphology-based detection of premature ventricular contractions. / Hadia, Rohit; Guldenring, Daniel; Finlay, Dewar; Kennedy, Alan; Janjua, Ghalib; Bond, Raymond; McLaughlin, James.

Unknown Host Publication. 2017.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Morphology-based detection of premature ventricular contractions

AU - Hadia, Rohit

AU - Guldenring, Daniel

AU - Finlay, Dewar

AU - Kennedy, Alan

AU - Janjua, Ghalib

AU - Bond, Raymond

AU - McLaughlin, James

PY - 2017/6/1

Y1 - 2017/6/1

N2 - Premature ventricular contraction (PVC) is the type of ectopic heartbeat, commonly found in the healthy population and is often considered benign. However, they are reported to adversely affect the accuracy of R-R variability based electrocardiographic (ECG) algorithms. This study proposes a Principal Component Analysis (PCA) based algorithmic approach to detect the PVCs based on their morphology. The eigenvectors were derived from signal window around the R-peak, where signal window for the PVC (wPVC) and that of NSR (wNSR) were set to 0.55 seconds and 0.16 seconds respectively. We used 24 ECG recordings from MIT BIH arrhythmia database as training dataset and the remaining 24 ECG recordings as testing dataset. Using the derived eigenvectors and the Linear regression (LR) analysis; complexes corresponding to the wNSR and wPVC were estimated from training and testing datasets. Four different classification methods were employed to differentiate between wPVS and wNSR, namely, Root mean squared error (RMSE), Pearson product-moment correlation coefficient comparision, Histogram probability distribution and k-Nearest Neighbour (KNN). All four methods were implemented individually to classify the wPVC and wNSR. The performance of each of the classification approach was evaluated by computing sensitivity and specificity. With the sensitivity of 93.45% and specificity of 93.14%, KNN based classification method has shown the best performance. The method proposed in this study allows for an effective differentiation between NSR beats and PVC beats.

AB - Premature ventricular contraction (PVC) is the type of ectopic heartbeat, commonly found in the healthy population and is often considered benign. However, they are reported to adversely affect the accuracy of R-R variability based electrocardiographic (ECG) algorithms. This study proposes a Principal Component Analysis (PCA) based algorithmic approach to detect the PVCs based on their morphology. The eigenvectors were derived from signal window around the R-peak, where signal window for the PVC (wPVC) and that of NSR (wNSR) were set to 0.55 seconds and 0.16 seconds respectively. We used 24 ECG recordings from MIT BIH arrhythmia database as training dataset and the remaining 24 ECG recordings as testing dataset. Using the derived eigenvectors and the Linear regression (LR) analysis; complexes corresponding to the wNSR and wPVC were estimated from training and testing datasets. Four different classification methods were employed to differentiate between wPVS and wNSR, namely, Root mean squared error (RMSE), Pearson product-moment correlation coefficient comparision, Histogram probability distribution and k-Nearest Neighbour (KNN). All four methods were implemented individually to classify the wPVC and wNSR. The performance of each of the classification approach was evaluated by computing sensitivity and specificity. With the sensitivity of 93.45% and specificity of 93.14%, KNN based classification method has shown the best performance. The method proposed in this study allows for an effective differentiation between NSR beats and PVC beats.

KW - Electrocardiography

KW - Principal component analysis

KW - Heart rate variability

KW - Morphology

KW - Training

KW - Testing

KW - Correlation coefficient

M3 - Conference contribution

SN - 978-1-5386-6630-2

BT - Unknown Host Publication

ER -

Hadia R, Guldenring D, Finlay D, Kennedy A, Janjua G, Bond R et al. Morphology-based detection of premature ventricular contractions. In Unknown Host Publication. 2017