Synthetic ECG signal generation using generative neural networks.

Electrocardiogram (ECG) datasets tend to be highly imbalanced due to the scarcity of abnormal cases. Additionally, the use of real patients' ECGs is highly regulated due to privacy issues. Therefore, there is always a need for more ECG data, especially for the training of automatic diagnosis ma...

Full description

Saved in:
Bibliographic Details
Main Authors: Edmond Adib, Fatemeh Afghah, John J Prevost
Format: Article
Language:English
Published: Public Library of Science (PLoS) 2025-01-01
Series:PLoS ONE
Online Access:https://doi.org/10.1371/journal.pone.0271270
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849769990652166144
author Edmond Adib
Fatemeh Afghah
John J Prevost
author_facet Edmond Adib
Fatemeh Afghah
John J Prevost
author_sort Edmond Adib
collection DOAJ
description Electrocardiogram (ECG) datasets tend to be highly imbalanced due to the scarcity of abnormal cases. Additionally, the use of real patients' ECGs is highly regulated due to privacy issues. Therefore, there is always a need for more ECG data, especially for the training of automatic diagnosis machine learning models, which perform better when trained on a balanced dataset. We studied the synthetic ECG generation capability of 5 different models from the generative adversarial network (GAN) family and compared their performances, the focus being only on Normal cardiac cycles. Dynamic Time Warping (DTW), Fréchet, and Euclidean distance functions were employed to quantitatively measure performance. Five different methods for evaluating generated beats were proposed and applied. We also proposed 3 new concepts (threshold, accepted beat and productivity rate) and employed them along with the aforementioned methods as a systematic way for comparison between models. The results show that all the tested models can, to an extent, successfully mass-generate acceptable heartbeats with high similarity in morphological features, and potentially all of them can be used to augment imbalanced datasets. However, visual inspections of generated beats favors BiLSTM-DC GAN and WGAN, as they produce statistically more acceptable beats. Also, with regards to productivity rate, the Classic GAN is superior with a 72% productivity rate. We also designed a simple experiment with the state-of-the-art classifier (ECGResNet34) to show empirically that the augmentation of the imbalanced dataset by synthetic ECG signals could improve the performance of classification significantly.
format Article
id doaj-art-bec08ac954a14fddb6930ce9edd8a748
institution DOAJ
issn 1932-6203
language English
publishDate 2025-01-01
publisher Public Library of Science (PLoS)
record_format Article
series PLoS ONE
spelling doaj-art-bec08ac954a14fddb6930ce9edd8a7482025-08-20T03:03:11ZengPublic Library of Science (PLoS)PLoS ONE1932-62032025-01-01203e027127010.1371/journal.pone.0271270Synthetic ECG signal generation using generative neural networks.Edmond AdibFatemeh AfghahJohn J PrevostElectrocardiogram (ECG) datasets tend to be highly imbalanced due to the scarcity of abnormal cases. Additionally, the use of real patients' ECGs is highly regulated due to privacy issues. Therefore, there is always a need for more ECG data, especially for the training of automatic diagnosis machine learning models, which perform better when trained on a balanced dataset. We studied the synthetic ECG generation capability of 5 different models from the generative adversarial network (GAN) family and compared their performances, the focus being only on Normal cardiac cycles. Dynamic Time Warping (DTW), Fréchet, and Euclidean distance functions were employed to quantitatively measure performance. Five different methods for evaluating generated beats were proposed and applied. We also proposed 3 new concepts (threshold, accepted beat and productivity rate) and employed them along with the aforementioned methods as a systematic way for comparison between models. The results show that all the tested models can, to an extent, successfully mass-generate acceptable heartbeats with high similarity in morphological features, and potentially all of them can be used to augment imbalanced datasets. However, visual inspections of generated beats favors BiLSTM-DC GAN and WGAN, as they produce statistically more acceptable beats. Also, with regards to productivity rate, the Classic GAN is superior with a 72% productivity rate. We also designed a simple experiment with the state-of-the-art classifier (ECGResNet34) to show empirically that the augmentation of the imbalanced dataset by synthetic ECG signals could improve the performance of classification significantly.https://doi.org/10.1371/journal.pone.0271270
spellingShingle Edmond Adib
Fatemeh Afghah
John J Prevost
Synthetic ECG signal generation using generative neural networks.
PLoS ONE
title Synthetic ECG signal generation using generative neural networks.
title_full Synthetic ECG signal generation using generative neural networks.
title_fullStr Synthetic ECG signal generation using generative neural networks.
title_full_unstemmed Synthetic ECG signal generation using generative neural networks.
title_short Synthetic ECG signal generation using generative neural networks.
title_sort synthetic ecg signal generation using generative neural networks
url https://doi.org/10.1371/journal.pone.0271270
work_keys_str_mv AT edmondadib syntheticecgsignalgenerationusinggenerativeneuralnetworks
AT fatemehafghah syntheticecgsignalgenerationusinggenerativeneuralnetworks
AT johnjprevost syntheticecgsignalgenerationusinggenerativeneuralnetworks