High-Accuracy Characterization of Ambulatory Holter Electrocardiogram Events: A Comparative Study Between Walsh-Hadamard Transform, First-Derivative-Based and Intelligent Techniques

High-Accuracy Characterization of Ambulatory Holter Electrocardiogram Events: A Comparative Study Between Walsh-Hadamard Transform, First-Derivative-Based and Intelligent Techniques

Mohammad Reza Homaeinezhad (Khajeh Nasir Toosi University of Technology, Tehran, Iran), Seyyed Amir Hoseini Sabzevari (Khajeh Nasir Toosi University of Technology, Tehran, Iran), Ali Ghaffari (Khajeh Nasir Toosi University of Technology, Tehran, Iran) and Mohammad Daevaeiha (Khajeh Nasir Toosi University of Technology and DAY General Hospital, Tehran, Iran)
DOI: 10.4018/ijsbbt.2012070102
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

In this paper, three noise-robust high-accuracy methods aiming at the detection and delineation of the electrocardiogram (ECG) events (QRS complex, P-wave, T-wave) were developed. The ECG signal was initially appropriately preprocessed by application of a bandpass FIR filter and Discrete Wavelet Transform (DWT). The first detection-delineation method was the Walsh-Hadamard Transform (WHT). The WHT coefficients were divided into two groups and the signal was reconstructed using the second group coefficients. By this reconstruction, the values of first derivative of events are made stronger rather than the values of other parts of signal. In the second method, a feed forward artificial neural network was implemented to detect all events of the ECG signal. In the third method, the first derivative of signal was computed using a new signal smoothing algorithm with corresponding statistical properties. For decreasing False Positive (FP) errors associated with P-wave detection, a discriminating border was introduced as the post processing stage specified by three QRS parameters: the duration of a QRS complex, the time distance from the former and latter QRS complexes, and the potential difference from former QRS complex J-location and the latter QRS complex fiducial location. The proposed methods were applied to DAY general hospital high resolution holter data.
Article Preview

Introduction

The electrocardiogram (ECG) signal is an electrical picture of the heart showing the effect of constitutive myocytes depolarization-repolarization waves propagation in the torso (Libby et al., 2007). The ECG signal can reflect the effect of many failures that occur in the heart either corresponding to the nervous or mechanical function of myocytes. The ECG is a diagnostic tool which is widely used in ICUs to monitor and assess patient’s heart long-term function(s). Some certain heart diseases such as T-Wave Alternans (Ghaffari et al., 2008; Ghaffari, Homaeinezhad, Akraminia, & Daevaeiha, 2009), Atrial Fibrillation (Ghaffari et al., 2009b; Chiarugi et al., 2007) and QT-prolongation (Christov et al., 2007) can be diagnosed by implementing statistical analyses on ECG signal. It is clear that detection and delineation of ECG events (QRS complex, P-wave, T-wave) is one of the main steps toward computerized analysis of the ECG signal. A large number of methods aimed for detection of ECG events have yet been proposed based on mathematical models (Sayadi et al., 2009), Hilbert transform and first derivative (Arzeno et al., 2008; Ghaffari, Homaeinezhad, Atarod, & Akraminia, 2010; Benitez et al., 2001), second order derivative (Mitra & Mitra, 2007), wavelet transform and filter banks (Ghaffari, Homaeinezhad, Khazraee, & Daevaeiha, 2010; Martinez et al., 2004; Ghaffari, Homaeinezhad, Akraminia, & Daevaeiha, 2009, 2010), soft computing (Neuro-fuzzy, genetic algorithm) (Kannathal et al., 2006), Hidden Markov Models (HMM) application (de Lannoy et al., 2008), etc. In this paper, three noise-robust high-accuracy methods aimed for detection and delineation of the ECG events based on Walsh-Hadamard Transform (WHT), Artificial Neural Network (ANN) and first derivative of ECG signal were developed. WHT is a linear orthogonal transformation that decomposes a signal into a set of rectangular or square waves with values of +1 or –1. WHT has following characteristics (Li et al., 2010):

  • 1)

    By increasing its root, the number of zero-crossing points is also increased, i.e., the frequency content is increased.

  • 2)

    The parity of Walsh function depends on the parity of its root.

  • 3)

    Because Walsh matrix has only +1 and -1, the multiplication can be used to perform addition and subtraction.

This paper is organized as follows: First (Materials and Methods) the method used in this study such as WHT, DWT, High-Pass Filtering (HPF), and smoothers are discussed. Then, the application of methods for the detection-delineation of ECG events including five subsections: two subsections for detection of QRS complex and three subsections for detection of P and T waves. We then (Result and Discussion) render the experimental results and comparing them with other studies. Finally, the conclusion from this study is presented.

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 5: 2 Issues (2017): Forthcoming, Available for Pre-Order
Volume 4: 2 Issues (2016): Forthcoming, Available for Pre-Order
Volume 3: 1 Issue (2015)
Volume 2: 4 Issues (2013)
Volume 1: 4 Issues (2012)
View Complete Journal Contents Listing