Automatic Detection of Atrial Fibrillation from Single-Lead ECG Using Deep Learning of the Cardiac Cycle

作     者：Alina Dubatovka Joachim M.Buhmann 

作者机构：Department of Computer ScienceETH ZurichZurichSwitzerland 

出 版 物：《Biomedical Engineering Frontiers》 (生物医学工程前沿（英文）)

年 卷 期：2022年第3卷第1期

页      面：159-170页

核心收录：

学科分类：1002[医学-临床医学] 100201[医学-内科学(含：心血管病、血液病、呼吸系病、消化系病、内分泌与代谢病、肾病、风湿病、传染病)] 10[医学] 

基　　金：funded by the Swiss Heart Failure Network (PHRT122/2018DRI14) (J.M.Buhmann PI) 

主　　题：recording Deep enable 

摘      要：Objective and Impact *** fibrillation(AF)is a serious medical condition that requires effective and timely treatment to prevent *** explore deep neural networks(DNNs)for learning cardiac cycles and reliably detecting AF from single-lead electrocardiogram(ECG)*** are widely used for diagnosis of various cardiac dysfunctions including *** huge amount of collected ECGs and recent algorithmic advances to process time-series data with DNNs substantially improve the accuracy of the AF ***,however,are often designed as general purpose black-box models and lack interpretability of their *** design a three-step pipeline for AF detection from ***,a recording is split into a sequence of individual heartbeats based on R-peak *** heartbeats are then encoded using a DNN that extracts interpretable features of a heartbeat by disentangling the duration of a heartbeat from its ***,the sequence of heartbeat codes is passed to a DNN to combine a signal-level representation capturing heart ***,the signal representations are passed to a DNN for detecting *** approach demonstrates a superior performance to existing ECG analysis methods on AF ***,the method provides interpretations of the features extracted from heartbeats by DNNs and enables cardiologists to study ECGs in terms of the shapes of individual heartbeats and rhythm of the whole *** considering ECGs on two levels and employing DNNs for modelling of cardiac cycles,this work presents a method for reliable detection of AF from single-lead ECGs.