6533b825fe1ef96bd1282652
RESEARCH PRODUCT
ECG Analysis for Ventricular Fibrillation Detection Using a Boltzmann Network
Juan F. Guerrero-martinezJose V. Frances-villoraManuel Bataller-mompeánAzeddine MjahadAlfredo Rosado-muñozsubject
medicine.medical_specialtybusiness.industryComputer scienceQuantitative Biology::Tissues and OrgansDetectorFeature extractionPattern recognitionmedicine.diseasesymbols.namesakeInternal medicineVentricular fibrillationBoltzmann constantmedicinesymbolsCardiologyPreprocessorECG analysisWaveformArtificial intelligencebusinessClassifier (UML)description
Arrhythmias consist on electrical alterations in the heart beat control. They can be identified by means of surface ECG leads. The main goal of this work is to provide a signal classification based on ECG signal waveform in the time-frequency domain especially targeted to Ventricular Fibrillation detection. The use of a classifier based on a Boltzmann network is proposed. However, a previous signal preprocessing is also required so that the Boltzmann network is fed with the appropriate data. In this case, an R-wave detector is used; after that, the Pseudo Wigner-Ville time-frequency distribution is obtained. This distribution is used to train and test the network, which handles it as an image and thus, provides a classification. Results show the ability of the network to provide a similar or higher classification ratio compared to other algorithms.
| year | journal | country | edition | language |
|---|---|---|---|---|
| 2015-01-01 |