Call For Paper Volume:4 Issue:8 Aug'2017 |

Combination of ECG Features with Artificial Neural Networks for the Detection of Ventricular Fibrillation

Publication Date : 29/06/2015



Author(s) :

KARTHIKA V S.


Volume/Issue :
Volume 2
,
Issue 6
(06 - 2015)



Abstract :

Early detection of cardiac pathologies is crucial for the success of the defibrillation therapy. A wide variety of detection algorithms have been proposed based on temporal, spectral or complexity parameters extracted from the ECG. However, these algorithms are mostly constructed by considering each parameter individually. This study presents a novel lifethreatening cardiac pathology detection algorithm that combines ECG parameters to a single feature vector and classifies using machine learning techniques. A total of 16 ECG parameters were computed accounting for morphological,spectral,complexity features and statistical measures of the ECG signal. A wavelet based feature extraction for statistical parameters was proposed to analyze, how they affect the detection performance. The proposed methodology was evaluated in the scenario,VF versus non-VF rhythms using the information contained in the medical imaging technology database. The combination of ECG parameters using statistical learning algorithms may improve the detection efficiency of life threatening cardiac pathologies.


No. of Downloads :

1


Indexing

Web Design MymensinghPremium WordPress ThemesWeb Development

Combination of ECG Features with Artificial Neural Networks for the Detection of Ventricular Fibrillation

June 29, 2015