Iranian Journal of Biomedical Engineering (IJBME)
نشریه علمی مهندسی پزشکی زیستی

Robust feature extracting from ECG signal using normalized capstral coefficients for heart arrhythmias recognition

Document Type : Full Research Paper

Authors

shadi aslani 1
Hossein Marvi 2
Mehdi Kafaee 3

1 MSC, Electronics, Shahrood University of technology, Shahrood, Iran

2 Associate Professor, Shahrood University of technology, Shahrood, Iran

3 Assistant Professor, Shahrood University of technology, Shahrood, Iran

https://doi.org/10.22041/ijbme.2025.2054411.1968
Abstract
Classification of electrocardiogram (ECG) signals is a crucial process in the diagnosis and treatment of cardiac abnormalities. ECG signals provide valuable information about the heart's condition, and classifying these signals as normal or abnormal plays a vital role in identifying and managing various cardiac disorders. In recent years, various feature extraction methods have been developed to enhance the accuracy and efficiency of automated ECG classification. This paper proposes a robust feature extraction method for ECG signals using normalized cepstral coefficients for the detection of cardiac arrhythmias. The extracted features in the proposed method are fed into a convolutional neural network (CNN) for classification. Additionally, various parameters of the CNN were examined to select the optimal network configuration. Statistical analysis demonstrates that the proposed feature extraction method and network achieve high accuracy in distinguishing unhealthy supraventricular arrhythmia and malignant ventricular ectopy signals from healthy signals. The experimental results indicate a 98% classification accuracy of the network in the testing phase.

Keywords

Cardiac Arrhythmia
Signal Processing
ECG Signal
CNN Network
Classification

Subjects

Iranian Journal of Biomedical Engineering (IJBME)
Volume 18, Issue 4
Winter 2025
Pages 399-408

  • Receive Date 25 February 2025
  • Revise Date 02 May 2025
  • Accept Date 15 May 2025

Home

Contact Us