Detail publikačního výsledku

ECG classification using neural networks and McSharry model

KIČMEROVÁ, D.; PROVAZNÍK, I.

Originální název

ECG classification using neural networks and McSharry model

Anglický název

ECG classification using neural networks and McSharry model

Druh

Stať ve sborníku v databázi WoS či Scopus

Originální abstrakt

The presented work is focused on modelling of arrhythmias using McSharry model followed by classification using an artificial neural network. The proposed method uses preprocessing of signals with Linear Approximation Distance Thresholding method and Line Segment Clustering method for establishing of initial parameters of McSharry model. The ECG data was taken from standard MIT/BIH arrhythmia database. Multilayer perceptron was used with classification accuracy of 90.1% for distinguishing of premature ventricular contraction and normal beat.

Anglický abstrakt

The presented work is focused on modelling of arrhythmias using McSharry model followed by classification using an artificial neural network. The proposed method uses preprocessing of signals with Linear Approximation Distance Thresholding method and Line Segment Clustering method for establishing of initial parameters of McSharry model. The ECG data was taken from standard MIT/BIH arrhythmia database. Multilayer perceptron was used with classification accuracy of 90.1% for distinguishing of premature ventricular contraction and normal beat.

Klíčová slova

ECG, McSharry model, neural networks

Klíčová slova v angličtině

ECG, McSharry model, neural networks

Autoři

KIČMEROVÁ, D.; PROVAZNÍK, I.

Vydáno

22.08.2007

Nakladatel

IEEE

Místo

Brno

ISBN

978-80-214-3409-7

Kniha

Proceedings of 5th IEEE Workshop Zvule

Strany od

1

Strany do

1

Strany počet

1

BibTex

@inproceedings{BUT22837,
  author="Dina {Kičmerová} and Valentýna {Provazník}",
  title="ECG classification using neural networks and McSharry model",
  booktitle="Proceedings of 5th IEEE Workshop Zvule",
  year="2007",
  pages="1--1",
  publisher="IEEE",
  address="Brno",
  isbn="978-80-214-3409-7"
}