Real-time discrimination of ventricular tachyarrhythmia with Fourier-transform neural network

Abstract

We have developed a method to discriminate life-threatening ventricular arrhythmias by observing the QRS complex of the electrocardiogram (ECG) in each heartbeat. Changes in QRS complexes due to rhythm origination and conduction path were observed with the Fourier transform, and three kinds of rhythms were discriminated by a neural network. In this paper, the potential of our method for clinical uses and real-time detection was examined using human surface ECG's and intracardiac electrograms (EGM's). The method achieved high sensitivity and specificity (> or = 0.98) in discrimination of supraventricular rhythms from ventricular ones. We also present a hardware implementation of the algorithm on a commercial single-chip CPU.

Pattern Classification Model Design and Performance Comparison for Data Mining of Time Series Data

“…심전도 신호 중 부정맥에 관한 파형의 측정분석은 파형의 크기 분석법 [16], 비선형 분석법 [14], 시주파수 분석법 [1], 신경회로망 [5,13,16,17], 퍼지추론 [16,17,19], SVM [5,16,17] [10,16] 3. 퍼지소속함수와 SVM …”

Development of the ‘Healthcor’ System as a Cardiac Disorders Symptoms Detector using an Expert System based on Arduino Uno

“…Some researchers have analyzed the frequency spectra of signals using Fourier Transform; unfortunately, this transform provides only the spectral information, and their temporal relationships are not included in the analysis. A representation that relates to time versus frequency information of the signal can be achieved using wavelets; this works well on nonstationary data (de Chazal et al, 2000;Minami et al, 1999;Romero & Serrano, 2001, Sarkaleh & Shahbahrami, 2012. Some algorithms have utilized heartbeat temporal intervals (Alexakis et al, 2003), morphological features (de Chazal et al, 2004), frequency domain features, and multi-fractal analysis (Ivanov et al, 2009).…”

Characterization of Ventricular Tachycardia and Fibrillation Using Semantic Mining

“…When artifial neural network (ANN) techniques have been proliferated in late 1990's, most researchers opt for this technique because it has the ability to deal with nonlinear discrimination between classes (Clayton et al, 1994;Minami et al, 1999). In addition, wavelet transforms (Khadra et al, 1997;Abbas et al, 2004;Nawarvar et al, 2004) and nonlinear analysis (Jekova et al, 2002;Sun et al, 2005;Daoming et al, 2007) were proposed as detection techniques for the classification of VF signals.…”

“…For the discrimination of VT and VF signals, the detection time becomes the key factor to determine the patient's fate. According to Minami et al 1999, the time delay due to the detection of VT/VF signals must be as short as possible; otherwise, the patient is at risk of death. If the proposed technique is too complex, the processing time will be too long to be effective.…”