Detection of cardiac arrest using deep learning for automated ext | 57642
Clinical & Experimental Cardiology

Clinical & Experimental Cardiology
Open Access

ISSN: 2155-9880

Detection of cardiac arrest using deep learning for automated external defibrillator

25th Annual Congress on Cardiology and Medical Interventions

July 16-17, 2018 | Atlanta, Georgia, USA

Minh Tuan Nguyen and Kiseon Kim

Gwangju Institute of Science and Technology, South Korea

Posters & Accepted Abstracts: J Clin Exp Cardiolog

Abstract :

Ventricular Fibrillation (VF) and Ventricular Tachycardia (VT), known as shockable rhythms, are the main cause of Sudden Cardiac Arrests (SCA), which can be properly treated by electronic defibrillators. Recently, Convolutional Neural Networks (CNN) has emerged as a useful approach. Moreover, CNNs are also used to extract a set of informative features, which is then used as the input of the Machine Learning (ML) classifiers. Here, we use two databases, which are the Creighton University Ventricular Tachyarrhythmia Database (CUDB) and the MIT-BIH Malignant Ventricular Arrhythmia Database (VFDB), divided into training and testing sets. Firstly, the CNNs and CNN-extractor models are selected on training data using fully CNNs and CNNs in combination with ML classifiers. Then, the fully CNNs and set of feature extracted by CNN-extractor are validated on the testing data. Therefore, the performance of fully CNNs and CNN-extractor is investigated to construct a novel algorithm, which can be effectively applied for the Automated External Defibrillators (AEDs).