Identification of Transient Noise to Reduce False Detections in Screening for Atrial Fibrillation
(2021) In Frontiers in Physiology 12.- Abstract
Screening for atrial fibrillation (AF) with a handheld device for recording the ECG is becoming increasingly popular. The poorer signal quality of such ECGs may lead to false detection of AF, often caused by transient noise. Consequently, the need for expert review in AF screening can become extensive. A convolutional neural network (CNN) is proposed for transient noise identification in AF detection. The network is trained using the events produced by a QRS detector, classified into either true beat detections or false detections. The CNN and a low-complexity AF detector are trained and tested using the StrokeStop I database, containing 30-s ECGs from mass screening for AF in the elderly population. Performance evaluation of the... (More)
Screening for atrial fibrillation (AF) with a handheld device for recording the ECG is becoming increasingly popular. The poorer signal quality of such ECGs may lead to false detection of AF, often caused by transient noise. Consequently, the need for expert review in AF screening can become extensive. A convolutional neural network (CNN) is proposed for transient noise identification in AF detection. The network is trained using the events produced by a QRS detector, classified into either true beat detections or false detections. The CNN and a low-complexity AF detector are trained and tested using the StrokeStop I database, containing 30-s ECGs from mass screening for AF in the elderly population. Performance evaluation of the CNN-based quality control using a subset of the database resulted in sensitivity, specificity, and accuracy of 96.4, 96.9, and 96.9%, respectively. By inserting the CNN before the AF detector, the false AF detections were reduced by 22.5% without any loss in sensitivity. The results show that the number of recordings calling for expert review can be significantly reduced thanks to the identification of transient noise. The reduction of false AF detections is directly linked to the time and cost spent on expert review.
(Less)
- author
- Halvaei, Hesam LU ; Svennberg, Emma ; Sörnmo, Leif LU and Stridh, Martin LU
- organization
- publishing date
- 2021
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- convolutional neural network, handheld devices, mass screening, short-term ECG signals, signal quality, transient noise
- in
- Frontiers in Physiology
- volume
- 12
- article number
- 672875
- publisher
- Frontiers Media S. A.
- external identifiers
-
- pmid:34149452
- scopus:85108220392
- ISSN
- 1664-042X
- DOI
- 10.3389/fphys.2021.672875
- language
- English
- LU publication?
- yes
- id
- 08e46b6a-c1ea-4b27-9698-8e9b19e93963
- date added to LUP
- 2021-07-15 14:34:55
- date last changed
- 2024-04-06 06:01:33
@article{08e46b6a-c1ea-4b27-9698-8e9b19e93963,
abstract = {{<p>Screening for atrial fibrillation (AF) with a handheld device for recording the ECG is becoming increasingly popular. The poorer signal quality of such ECGs may lead to false detection of AF, often caused by transient noise. Consequently, the need for expert review in AF screening can become extensive. A convolutional neural network (CNN) is proposed for transient noise identification in AF detection. The network is trained using the events produced by a QRS detector, classified into either true beat detections or false detections. The CNN and a low-complexity AF detector are trained and tested using the StrokeStop I database, containing 30-s ECGs from mass screening for AF in the elderly population. Performance evaluation of the CNN-based quality control using a subset of the database resulted in sensitivity, specificity, and accuracy of 96.4, 96.9, and 96.9%, respectively. By inserting the CNN before the AF detector, the false AF detections were reduced by 22.5% without any loss in sensitivity. The results show that the number of recordings calling for expert review can be significantly reduced thanks to the identification of transient noise. The reduction of false AF detections is directly linked to the time and cost spent on expert review.</p>}},
author = {{Halvaei, Hesam and Svennberg, Emma and Sörnmo, Leif and Stridh, Martin}},
issn = {{1664-042X}},
keywords = {{convolutional neural network; handheld devices; mass screening; short-term ECG signals; signal quality; transient noise}},
language = {{eng}},
publisher = {{Frontiers Media S. A.}},
series = {{Frontiers in Physiology}},
title = {{Identification of Transient Noise to Reduce False Detections in Screening for Atrial Fibrillation}},
url = {{https://lup.lub.lu.se/search/files/117084867/fphys_12_672875.pdf}},
doi = {{10.3389/fphys.2021.672875}},
volume = {{12}},
year = {{2021}},
}