Training Convolutional Neural Networks on Simulated Photoplethysmography Data : Application to Bradycardia and Tachycardia Detection
(2022) In Frontiers in Physiology 13.- Abstract
Objective: To develop a method for detection of bradycardia and ventricular tachycardia using the photoplethysmogram (PPG). Approach: The detector is based on a dual-branch convolutional neural network (CNN), whose input is the scalograms of the continuous wavelet transform computed in 5-s segments. Training and validation of the CNN is accomplished using simulated PPG signals generated from RR interval series extracted from public ECG databases. Manually annotated real PPG signals from the PhysioNet/CinC 2015 Challenge Database are used for performance evaluation. The performance is compared to that of a pulse-based reference detector. Results: The sensitivity/specificity were found to be 98.1%/97.9 and 76.6%/96.8% for the CNN-based... (More)
Objective: To develop a method for detection of bradycardia and ventricular tachycardia using the photoplethysmogram (PPG). Approach: The detector is based on a dual-branch convolutional neural network (CNN), whose input is the scalograms of the continuous wavelet transform computed in 5-s segments. Training and validation of the CNN is accomplished using simulated PPG signals generated from RR interval series extracted from public ECG databases. Manually annotated real PPG signals from the PhysioNet/CinC 2015 Challenge Database are used for performance evaluation. The performance is compared to that of a pulse-based reference detector. Results: The sensitivity/specificity were found to be 98.1%/97.9 and 76.6%/96.8% for the CNN-based detector, respectively, whereas the corresponding results for the pulse-based detector were 94.7%/99.8 and 67.1%/93.8%, respectively. Significance: The proposed detector may be useful for continuous, long-term monitoring of bradycardia and tachycardia using wearable devices, e.g., wrist-worn devices, especially in situations where sensitivity is favored over specificity. The study demonstrates that simulated PPG signals are suitable for training and validation of a CNN.
(Less)
- author
- Sološenko, Andrius ; Paliakaitė, Birutė ; Marozas, Vaidotas and Sörnmo, Leif LU
- organization
- publishing date
- 2022
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- bradycardia, convolutional neural networks, detection, photoplethysmogram, simulated signals, tachycardia
- in
- Frontiers in Physiology
- volume
- 13
- article number
- 928098
- publisher
- Frontiers Media S. A.
- external identifiers
-
- pmid:35923223
- scopus:85135598738
- ISSN
- 1664-042X
- DOI
- 10.3389/fphys.2022.928098
- language
- English
- LU publication?
- yes
- id
- bcfcb14f-76d3-443f-bb3b-b3f20146a24b
- date added to LUP
- 2022-09-19 10:15:39
- date last changed
- 2025-03-07 17:58:50
@article{bcfcb14f-76d3-443f-bb3b-b3f20146a24b,
abstract = {{<p>Objective: To develop a method for detection of bradycardia and ventricular tachycardia using the photoplethysmogram (PPG). Approach: The detector is based on a dual-branch convolutional neural network (CNN), whose input is the scalograms of the continuous wavelet transform computed in 5-s segments. Training and validation of the CNN is accomplished using simulated PPG signals generated from RR interval series extracted from public ECG databases. Manually annotated real PPG signals from the PhysioNet/CinC 2015 Challenge Database are used for performance evaluation. The performance is compared to that of a pulse-based reference detector. Results: The sensitivity/specificity were found to be 98.1%/97.9 and 76.6%/96.8% for the CNN-based detector, respectively, whereas the corresponding results for the pulse-based detector were 94.7%/99.8 and 67.1%/93.8%, respectively. Significance: The proposed detector may be useful for continuous, long-term monitoring of bradycardia and tachycardia using wearable devices, e.g., wrist-worn devices, especially in situations where sensitivity is favored over specificity. The study demonstrates that simulated PPG signals are suitable for training and validation of a CNN.</p>}},
author = {{Sološenko, Andrius and Paliakaitė, Birutė and Marozas, Vaidotas and Sörnmo, Leif}},
issn = {{1664-042X}},
keywords = {{bradycardia; convolutional neural networks; detection; photoplethysmogram; simulated signals; tachycardia}},
language = {{eng}},
publisher = {{Frontiers Media S. A.}},
series = {{Frontiers in Physiology}},
title = {{Training Convolutional Neural Networks on Simulated Photoplethysmography Data : Application to Bradycardia and Tachycardia Detection}},
url = {{http://dx.doi.org/10.3389/fphys.2022.928098}},
doi = {{10.3389/fphys.2022.928098}},
volume = {{13}},
year = {{2022}},
}