Automated identification of shockable and non-shockable life-threatening ventricular arrhythmias using convolutional neural network

被引:200
作者
Acharya, U. Rajendra [1 ,2 ,3 ]
Fujita, Hamido [4 ]
Oh, Shu Lih [1 ]
Raghavendra, U. [5 ]
Tan, Jen Hong [1 ]
Adam, Muhammad [1 ]
Gertych, Arkadiusz [6 ]
Hagiwara, Yuki [1 ]
机构
[1] Ngee Ann Polytech, Dept Elect & Comp Engn, Singapore, Singapore
[2] Singapore Univ Social Sci, Sch Sci & Technol, Dept Biomed Engn, Singapore, Singapore
[3] Univ Malaya, Fac Engn, Dept Biomed Engn, Kuala Lumpur, Malaysia
[4] Iwate Prefectural Univ, Fac Software & Informat Sci, Takizawa, Iwate 0200693, Japan
[5] Manipal Univ, Manipal Inst Technol, Dept Instrumentat & Control Engn, Manipal 576104, India
[6] Cedars Sinai Med Ctr, Dept Pathol & Lab Med, Dept Surg, Los Angeles, CA 90048 USA
来源
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE | 2018年 / 79卷
关键词
Automated external defibrillator (AED); ECG signals; Non-shockable; Shockable; Ventricular arrhythmias; RESUSCITATION COUNCIL GUIDELINES; FIBRILLATION; DEFIBRILLATORS; MANAGEMENT;
D O I
10.1016/j.future.2017.08.039
中图分类号
TP301 [理论、方法];
学科分类号
080201 [机械制造及其自动化];
摘要
Ventricular tachycardia (VT) and ventricular fibrillation (VFib) are the life-threatening shockable arrhythmias which require immediate attention. Cardiopulmonary resuscitation (CPR) and defibrillation are highly recommended means of immediate treatment of these shockable arrhythmias and to resume spontaneous circulation. However, to increase efficacy of defibrillation by an automated external defibrillator (AED), an accurate distinction of shockable ventricular arrhythmias from non-shockable ones needs to be provided upfront. Therefore, in this work, we have proposed a novel tool for an automated differentiation of shockable and non-shockable ventricular arrhythmias from 2 s electrocardiogram (ECG) segments. Segmented ECGs are processed by an eleven-layer convolutional neural network (CNN) model. Our proposed system was 10-fold cross validated and achieved maximum accuracy, sensitivity and specificity of 93.18%, 95.32% and 91.04% respectively. Its high performance indicates that shockable life-threatening arrhythmia can be immediately detected and thus increase the chance of survival while CPR or AED-based support is performed. Our tool can also be seamlessly integrated with an ECG acquisition systems in the intensive care units (ICUs). (C) 2017 Elsevier B.V. All rights reserved.
引用
收藏
页码:952 / 959
页数:8
相关论文
共 41 条
[1]
Study of heart rate variability signals at sitting and lying postures [J].
Acharya, Rajendra U. ;
Kannathal, N. ;
Hua, Lee Mei ;
Yi, Leong Mei .
JOURNAL OF BODYWORK AND MOVEMENT THERAPIES, 2005, 9 (02) :134-141
[2]
Acharya U. R., 2017, INFORM SCI NY, V405
[3]
Acharya U. R., 2017, KNOWL-BASED SYST, V946, P1
[4]
Application of deep convolutional neural network for automated detection of myocardial infarction using ECG signals [J].
Acharya, U. Rajendra ;
Fujita, Hamido ;
Oh, Shu Lih ;
Hagiwara, Yuki ;
Tan, Jen Hong ;
Adam, Muhammad .
INFORMATION SCIENCES, 2017, 415 :190-198
[5]
Detection of Life-Threatening Arrhythmias Using Feature Selection and Support Vector Machines [J].
Alonso-Atienza, Felipe ;
Morgado, Eduardo ;
Fernandez-Martinez, Lorena ;
Garcia-Alberola, Arcadi ;
Luis Rojo-Alvarez, Jose .
IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, 2014, 61 (03) :832-840
[6]
Reliability of old and new ventricular fibrillation detection algorithms for automated external defibrillators [J].
Amann, Anton ;
Tratnig, Robert ;
Unterkofler, Karl .
BIOMEDICAL ENGINEERING ONLINE, 2005, 4 (1)
[7]
Detecting ventricular fibrillation by time-delay methods [J].
Amann, Anton ;
Tratnig, Robert ;
Unterkofler, Karl .
IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, 2007, 54 (01) :174-177
[8]
[Anonymous], 2016, MAYO CLIN STAFF HEAR
[9]
[Anonymous], CONV NET NETW VIS RE
[10]
[Anonymous], 2017, COMPUTERS BIOL MED