Non-invasive Estimation of Atrial Fibrillation Driver Position With Convolutional Neural Networks and Body Surface Potentials

被引:7
作者
Camara-Vazquez, Miguel Angel [1 ]
Hernandez-Romero, Ismael [1 ]
Morgado-Reyes, Eduardo [1 ]
Guillem, Maria S. [2 ]
Climent, Andreu M. [2 ]
Barquero-Perez, Oscar [1 ,2 ]
机构
[1] Rey Juan Carlos Univ, Telemat Syst & Computat, Dept Signal Theory & Commun, Madrid, Spain
[2] Univ Politecn Valencia, ITACA Inst, Valencia, Spain
关键词
atrial fibrillation; body surface potentials; driver position; convolutional neural networks; deep learning; CONVENTIONAL ABLATION; INVERSE PROBLEMS; FOCAL IMPULSE; REGULARIZATION; MANAGEMENT; ROTORS;
D O I
10.3389/fphys.2021.733449
中图分类号
Q4 [生理学];
学科分类号
071003 ;
摘要
Atrial fibrillation (AF) is characterized by complex and irregular propagation patterns, and AF onset locations and drivers responsible for its perpetuation are the main targets for ablation procedures. ECG imaging (ECGI) has been demonstrated as a promising tool to identify AF drivers and guide ablation procedures, being able to reconstruct the electrophysiological activity on the heart surface by using a non-invasive recording of body surface potentials (BSP). However, the inverse problem of ECGI is ill-posed, and it requires accurate mathematical modeling of both atria and torso, mainly from CT or MR images. Several deep learning-based methods have been proposed to detect AF, but most of the AF-based studies do not include the estimation of ablation targets. In this study, we propose to model the location of AF drivers from BSP as a supervised classification problem using convolutional neural networks (CNN). Accuracy in the test set ranged between 0.75 (SNR = 5 dB) and 0.93 (SNR = 20 dB upward) when assuming time independence, but it worsened to 0.52 or lower when dividing AF models into blocks. Therefore, CNN could be a robust method that could help to non-invasively identify target regions for ablation in AF by using body surface potential mapping, avoiding the use of ECGI.
引用
收藏
页数:11
相关论文
共 48 条
[1]   Atrial Fibrillation Driver Localization From Body Surface Potentials Using Deep Learning [J].
Angel Camara-Vazquez, Miguel ;
Oter-Astillero, Adrian ;
Hernandez-Romero, Ismael ;
Rodrigo, Miguel ;
Morgado-Reyes, Eduardo ;
Guillem Sanchez, Maria de la Salud ;
Climent, Andreu M. ;
Barquero-Perez, Oscar .
2020 COMPUTING IN CARDIOLOGY, 2020,
[2]   RELATING EPICARDIAL TO BODY-SURFACE POTENTIAL DISTRIBUTIONS BY MEANS OF TRANSFER-COEFFICIENTS BASED ON GEOMETRY MEASUREMENTS [J].
BARR, RC ;
RAMSEY, M ;
SPACH, MS .
IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, 1977, 24 (01) :1-11
[3]   Electrical imaging of the heart [J].
Brooks, DH ;
MacLeod, RS .
IEEE SIGNAL PROCESSING MAGAZINE, 1997, 14 (01) :24-42
[4]   CT-Scan Free Neural Network-Based Reconstruction of Heart Surface Potentials From ECG Recordings [J].
Bujnarowski, Kamil ;
Bonizzi, Pietro ;
Cluitmans, Matthijs ;
Peeters, Ralf ;
Karel, Joel .
2020 COMPUTING IN CARDIOLOGY, 2020,
[5]   Xception: Deep Learning with Depthwise Separable Convolutions [J].
Chollet, Francois .
30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), 2017, :1800-1807
[6]   Worldwide Epidemiology of Atrial Fibrillation A Global Burden of Disease 2010 Study [J].
Chugh, Sumeet S. ;
Havmoeller, Rasmus ;
Narayanan, Kumar ;
Singh, David ;
Rienstra, Michiel ;
Benjamin, Emelia J. ;
Gillum, Richard F. ;
Kim, Young-Hoon ;
McAnulty, John H. ;
Zheng, Zhi-Jie ;
Forouzanfar, Mohammad H. ;
Naghavi, Mohsen ;
Mensah, George A. ;
Ezzati, Majid ;
Murray, Christopher J. L. .
CIRCULATION, 2014, 129 (08) :837-847
[7]   Noninvasive Characterization of Epicardial Activation in Humans With Diverse Atrial Fibrillation Patterns [J].
Cuculich, Phillip S. ;
Wang, Yong ;
Lindsay, Bruce D. ;
Faddis, Mitchell N. ;
Schuessler, Richard B. ;
Damiano, Ralph J., Jr. ;
Li, Li ;
Rudy, Yoram .
CIRCULATION, 2010, 122 (14) :1364-+
[8]   Diagnosis-to-ablation time as a predictor of success: early choice for pulmonary vein isolation and long-term outcome in atrial fibrillation: results from the Middelheim-PVI Registry [J].
De Greef, Y. ;
Schwagten, B. ;
Chierchia, G. B. ;
de Asmundis, C. ;
Stockman, D. ;
Buysschaert, I. .
EUROPACE, 2018, 20 (04) :589-595
[9]   A LINEAR DISCRETIZATION OF THE VOLUME CONDUCTOR BOUNDARY INTEGRAL-EQUATION USING ANALYTICALLY INTEGRATED ELEMENTS [J].
DEMUNCK, JC .
IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, 1992, 39 (09) :986-990
[10]   Computational modeling of the human atrial anatomy and electrophysiology [J].
Doessel, Olaf ;
Krueger, Martin W. ;
Weber, Frank M. ;
Wilhelms, Mathias ;
Seemann, Gunnar .
MEDICAL & BIOLOGICAL ENGINEERING & COMPUTING, 2012, 50 (08) :773-799