Cargando…

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

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 abl...

Descripción completa

Detalles Bibliográficos
Autores principales: Cámara-Vázquez, Miguel Ángel, Hernández-Romero, Ismael, Morgado-Reyes, Eduardo, Guillem, Maria S., Climent, Andreu M., Barquero-Pérez, Oscar
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8552066/
https://www.ncbi.nlm.nih.gov/pubmed/34721065
http://dx.doi.org/10.3389/fphys.2021.733449
_version_ 1784591304063188992
author Cámara-Vázquez, Miguel Ángel
Hernández-Romero, Ismael
Morgado-Reyes, Eduardo
Guillem, Maria S.
Climent, Andreu M.
Barquero-Pérez, Oscar
author_facet Cámara-Vázquez, Miguel Ángel
Hernández-Romero, Ismael
Morgado-Reyes, Eduardo
Guillem, Maria S.
Climent, Andreu M.
Barquero-Pérez, Oscar
author_sort Cámara-Vázquez, Miguel Ángel
collection PubMed
description 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.
format Online
Article
Text
id pubmed-8552066
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher Frontiers Media S.A.
record_format MEDLINE/PubMed
spelling pubmed-85520662021-10-29 Non-invasive Estimation of Atrial Fibrillation Driver Position With Convolutional Neural Networks and Body Surface Potentials Cámara-Vázquez, Miguel Ángel Hernández-Romero, Ismael Morgado-Reyes, Eduardo Guillem, Maria S. Climent, Andreu M. Barquero-Pérez, Oscar Front Physiol Physiology 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. Frontiers Media S.A. 2021-10-14 /pmc/articles/PMC8552066/ /pubmed/34721065 http://dx.doi.org/10.3389/fphys.2021.733449 Text en Copyright © 2021 Cámara-Vázquez, Hernández-Romero, Morgado-Reyes, Guillem, Climent and Barquero-Pérez. https://creativecommons.org/licenses/by/4.0/This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
spellingShingle Physiology
Cámara-Vázquez, Miguel Ángel
Hernández-Romero, Ismael
Morgado-Reyes, Eduardo
Guillem, Maria S.
Climent, Andreu M.
Barquero-Pérez, Oscar
Non-invasive Estimation of Atrial Fibrillation Driver Position With Convolutional Neural Networks and Body Surface Potentials
title Non-invasive Estimation of Atrial Fibrillation Driver Position With Convolutional Neural Networks and Body Surface Potentials
title_full Non-invasive Estimation of Atrial Fibrillation Driver Position With Convolutional Neural Networks and Body Surface Potentials
title_fullStr Non-invasive Estimation of Atrial Fibrillation Driver Position With Convolutional Neural Networks and Body Surface Potentials
title_full_unstemmed Non-invasive Estimation of Atrial Fibrillation Driver Position With Convolutional Neural Networks and Body Surface Potentials
title_short Non-invasive Estimation of Atrial Fibrillation Driver Position With Convolutional Neural Networks and Body Surface Potentials
title_sort non-invasive estimation of atrial fibrillation driver position with convolutional neural networks and body surface potentials
topic Physiology
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8552066/
https://www.ncbi.nlm.nih.gov/pubmed/34721065
http://dx.doi.org/10.3389/fphys.2021.733449
work_keys_str_mv AT camaravazquezmiguelangel noninvasiveestimationofatrialfibrillationdriverpositionwithconvolutionalneuralnetworksandbodysurfacepotentials
AT hernandezromeroismael noninvasiveestimationofatrialfibrillationdriverpositionwithconvolutionalneuralnetworksandbodysurfacepotentials
AT morgadoreyeseduardo noninvasiveestimationofatrialfibrillationdriverpositionwithconvolutionalneuralnetworksandbodysurfacepotentials
AT guillemmarias noninvasiveestimationofatrialfibrillationdriverpositionwithconvolutionalneuralnetworksandbodysurfacepotentials
AT climentandreum noninvasiveestimationofatrialfibrillationdriverpositionwithconvolutionalneuralnetworksandbodysurfacepotentials
AT barqueroperezoscar noninvasiveestimationofatrialfibrillationdriverpositionwithconvolutionalneuralnetworksandbodysurfacepotentials