Detection of Brief Episodes of Atrial Fibrillation Based on Electrocardiomatrix and Convolutional Neural Network
(2021) In Frontiers in Physiology 12. p.1-16- Abstract
- Background: Brief episodes of atrial fibrillation (AF) may evolve into longer AF episodes increasing the chances of thrombus formation, stroke, and death. Classical methods for AF detection investigate rhythm irregularity or P-wave absence in the ECG, while deep learning approaches profit from the availability of annotated ECG databases to learn discriminatory features linked to different diagnosis. However, some deep learning approaches do not provide analysis of the features used for classification. This paper introduces a convolutional neural network (CNN) approach for automatic detection of brief AF episodes based on electrocardiomatrix-images (ECM-images) aiming to link deep learning to features with clinical... (More)
- Background: Brief episodes of atrial fibrillation (AF) may evolve into longer AF episodes increasing the chances of thrombus formation, stroke, and death. Classical methods for AF detection investigate rhythm irregularity or P-wave absence in the ECG, while deep learning approaches profit from the availability of annotated ECG databases to learn discriminatory features linked to different diagnosis. However, some deep learning approaches do not provide analysis of the features used for classification. This paper introduces a convolutional neural network (CNN) approach for automatic detection of brief AF episodes based on electrocardiomatrix-images (ECM-images) aiming to link deep learning to features with clinical meaning.
Materials and Methods: The CNN is trained using two databases: the Long-Term Atrial Fibrillation and the MIT-BIH Normal Sinus Rhythm, and tested on three databases: the MIT-BIH Atrial Fibrillation, the MIT-BIH Arrhythmia, and the Monzino-AF. Detection of AF is done using a sliding window of 10 beats plus 3 s. Performance is quantified using both standard classification metrics and the EC57 standard for arrhythmia detection. Layer-wise relevance propagation analysis was applied to link the decisions made by the CNN to clinical characteristics in the ECG.
Results: For all three testing databases, episode sensitivity was greater than 80.22, 89.66, and 97.45% for AF episodes shorter than 15, 30 s, and for all episodes, respectively.
Conclusions: Rhythm and morphological characteristics of the electrocardiogram can be learned by a CNN from ECM-images for the detection of brief episodes of AF. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/ef84b229-e075-43ee-868d-12a21d7468ca
- author
- Salinas Martinez, Ricardo LU ; de Bie, Johannes ; Marzocchi, Nicoletta and Sandberg, Frida LU
- organization
- publishing date
- 2021-08-25
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- Atrial Fibrillation, brief atrial fibrillation, Convolutional neural network, interpretability, atrial fibrillation detection, layer-wise relevance propagation, long-term ECG
- in
- Frontiers in Physiology
- volume
- 12
- article number
- 673819
- pages
- 16 pages
- publisher
- Frontiers Media S. A.
- external identifiers
-
- scopus:85114641048
- pmid:34512372
- ISSN
- 1664-042X
- DOI
- 10.3389/fphys.2021.673819
- project
- Ph.D. project: Paroxysmal atrial fibrillation: Continuous tracking of arrhythmia progression
- MultidisciplinarY training network for ATrial fibRillation monItoring, treAtment and progression
- language
- English
- LU publication?
- yes
- id
- ef84b229-e075-43ee-868d-12a21d7468ca
- date added to LUP
- 2021-09-29 15:31:45
- date last changed
- 2023-06-21 10:11:07
@article{ef84b229-e075-43ee-868d-12a21d7468ca,
abstract = {{<b>Background:</b> Brief episodes of atrial fibrillation (AF) may evolve into longer AF episodes increasing the chances of thrombus formation, stroke, and death. Classical methods for AF detection investigate rhythm irregularity or P-wave absence in the ECG, while deep learning approaches profit from the availability of annotated ECG databases to learn discriminatory features linked to different diagnosis. However, some deep learning approaches do not provide analysis of the features used for classification. This paper introduces a convolutional neural network (CNN) approach for automatic detection of brief AF episodes based on electrocardiomatrix-images (ECM-images) aiming to link deep learning to features with clinical meaning.<br/><b>Materials and Methods:</b> The CNN is trained using two databases: the Long-Term Atrial Fibrillation and the MIT-BIH Normal Sinus Rhythm, and tested on three databases: the MIT-BIH Atrial Fibrillation, the MIT-BIH Arrhythmia, and the Monzino-AF. Detection of AF is done using a sliding window of 10 beats plus 3 s. Performance is quantified using both standard classification metrics and the EC57 standard for arrhythmia detection. Layer-wise relevance propagation analysis was applied to link the decisions made by the CNN to clinical characteristics in the ECG.<br/><b>Results: </b>For all three testing databases, episode sensitivity was greater than 80.22, 89.66, and 97.45% for AF episodes shorter than 15, 30 s, and for all episodes, respectively.<br/><b>Conclusions:</b> Rhythm and morphological characteristics of the electrocardiogram can be learned by a CNN from ECM-images for the detection of brief episodes of AF.}},
author = {{Salinas Martinez, Ricardo and de Bie, Johannes and Marzocchi, Nicoletta and Sandberg, Frida}},
issn = {{1664-042X}},
keywords = {{Atrial Fibrillation; brief atrial fibrillation; Convolutional neural network; interpretability; atrial fibrillation detection; layer-wise relevance propagation; long-term ECG}},
language = {{eng}},
month = {{08}},
pages = {{1--16}},
publisher = {{Frontiers Media S. A.}},
series = {{Frontiers in Physiology}},
title = {{Detection of Brief Episodes of Atrial Fibrillation Based on Electrocardiomatrix and Convolutional Neural Network}},
url = {{http://dx.doi.org/10.3389/fphys.2021.673819}},
doi = {{10.3389/fphys.2021.673819}},
volume = {{12}},
year = {{2021}},
}