An expert system for detection of electrocardiographic changes in patients with partial epilepsy using wavelet-based neural networks

Guler İ., Ubeyli E.

EXPERT SYSTEMS, vol.22, no.2, pp.62-71, 2005 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 22 Issue: 2
  • Publication Date: 2005
  • Doi Number: 10.1111/j.1468-0394.2005.00295.x
  • Journal Name: EXPERT SYSTEMS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.62-71
  • Keywords: electrocardiographic changes, multilayer perceptron neural networks, wavelet transform, training algorithms, partial epilepsy, ECG PATTERN-RECOGNITION, HEART-RATE, CLASSIFICATION, SEIZURES, SIGNALS, TRANSFORMS, STENOSIS
  • Gazi University Affiliated: Yes


In this study a wavelet-based neural network model, employing the multilayer perceptron, is presented for the detection of electrocardiographic changes in patients with partial epilepsy. Decision making is performed in two stages: feature extraction using the wavelet transform, and multilayer perceptron neural networks (MLPNNs) trained with the backpropagation, delta-bar-delta, extended delta-bar-delta and quick propagation algorithms as classifiers. The classification results, the values of statistical parameters and performance evaluation parameters of the MLPNNs trained with different algorithms are compared. Two types of electrocardiogram beats (normal and partial epilepsy) obtained from the MIT-BIH database were classified with accuracy varying from 90.00% to 97.50% by the MLPNNs trained with different algorithms.