Generation of Seismocardiography Heartbeats Using a Wasserstein Generative Adversarial Network With Feature Control
<italic>Goal:</italic> Seismocardiography (SCG) offers critical insights into cardiac performance, but its analysis often faces challenges due to the limited availability of data. This study aims to generate synthetic SCG heartbeats which can augment existing datasets to enable more rese...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Open Journal of Engineering in Medicine and Biology |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10731564/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846162927021719552 |
|---|---|
| author | James Skoric Yannick D'Mello David V. Plant |
| author_facet | James Skoric Yannick D'Mello David V. Plant |
| author_sort | James Skoric |
| collection | DOAJ |
| description | <italic>Goal:</italic> Seismocardiography (SCG) offers critical insights into cardiac performance, but its analysis often faces challenges due to the limited availability of data. This study aims to generate synthetic SCG heartbeats which can augment existing datasets to enable more research avenues. <italic>Methods</italic>: We trained a Wasserstein generative adversarial network (GAN) with gradient penalty on authentic SCG heartbeats. It was conditioned with embedded subject-specific identifiers to create individualized heartbeats. We employed linear permutations in the latent and conditional spaces to control signal features, and a convolutional network to classify lung volume states from real and synthetic data separately. <italic>Results</italic>: The model effectively replicated SCG signal morphology, while maintaining a level of variance which matches the variability of cardiac activity. Comparisons with real SCG waveforms yielded Pearson's r-squared correlation of 0.62 for average heartbeats. Linear manipulations were successful in controlling simple features although they were limited in more complex characteristics. Additionally, the model demonstrated strong performance in practical applications, with the synthetic data achieving an accuracy of 88% in lung volume classification as compared to 89% achieved with real data. Augmenting real data with additional synthetic data improved performance by 3%. <italic>Conclusions</italic>: GANs for artificial SCG heartbeat generation produce realistic and diverse results that have the potential to overcome data limitations, thereby enhancing SCG-based research. |
| format | Article |
| id | doaj-art-b5db651ea9af4a82a7dcc84ccc07f4a0 |
| institution | Kabale University |
| issn | 2644-1276 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Open Journal of Engineering in Medicine and Biology |
| spelling | doaj-art-b5db651ea9af4a82a7dcc84ccc07f4a02024-11-20T00:01:39ZengIEEEIEEE Open Journal of Engineering in Medicine and Biology2644-12762025-01-01611912610.1109/OJEMB.2024.348553510731564Generation of Seismocardiography Heartbeats Using a Wasserstein Generative Adversarial Network With Feature ControlJames Skoric0https://orcid.org/0000-0003-3418-6635Yannick D'Mello1https://orcid.org/0000-0001-7922-7475David V. Plant2Department of Electrical and Computer Engineering, McGill University, Montreal, QC, CanadaDepartment of Electrical and Computer Engineering, McGill University, Montreal, QC, CanadaDepartment of Electrical and Computer Engineering, McGill University, Montreal, QC, Canada<italic>Goal:</italic> Seismocardiography (SCG) offers critical insights into cardiac performance, but its analysis often faces challenges due to the limited availability of data. This study aims to generate synthetic SCG heartbeats which can augment existing datasets to enable more research avenues. <italic>Methods</italic>: We trained a Wasserstein generative adversarial network (GAN) with gradient penalty on authentic SCG heartbeats. It was conditioned with embedded subject-specific identifiers to create individualized heartbeats. We employed linear permutations in the latent and conditional spaces to control signal features, and a convolutional network to classify lung volume states from real and synthetic data separately. <italic>Results</italic>: The model effectively replicated SCG signal morphology, while maintaining a level of variance which matches the variability of cardiac activity. Comparisons with real SCG waveforms yielded Pearson's r-squared correlation of 0.62 for average heartbeats. Linear manipulations were successful in controlling simple features although they were limited in more complex characteristics. Additionally, the model demonstrated strong performance in practical applications, with the synthetic data achieving an accuracy of 88% in lung volume classification as compared to 89% achieved with real data. Augmenting real data with additional synthetic data improved performance by 3%. <italic>Conclusions</italic>: GANs for artificial SCG heartbeat generation produce realistic and diverse results that have the potential to overcome data limitations, thereby enhancing SCG-based research.https://ieeexplore.ieee.org/document/10731564/Cardiovascular monitoringgenerative adversarial networksseismocardiographysynthetic generationwearable monitoring |
| spellingShingle | James Skoric Yannick D'Mello David V. Plant Generation of Seismocardiography Heartbeats Using a Wasserstein Generative Adversarial Network With Feature Control IEEE Open Journal of Engineering in Medicine and Biology Cardiovascular monitoring generative adversarial networks seismocardiography synthetic generation wearable monitoring |
| title | Generation of Seismocardiography Heartbeats Using a Wasserstein Generative Adversarial Network With Feature Control |
| title_full | Generation of Seismocardiography Heartbeats Using a Wasserstein Generative Adversarial Network With Feature Control |
| title_fullStr | Generation of Seismocardiography Heartbeats Using a Wasserstein Generative Adversarial Network With Feature Control |
| title_full_unstemmed | Generation of Seismocardiography Heartbeats Using a Wasserstein Generative Adversarial Network With Feature Control |
| title_short | Generation of Seismocardiography Heartbeats Using a Wasserstein Generative Adversarial Network With Feature Control |
| title_sort | generation of seismocardiography heartbeats using a wasserstein generative adversarial network with feature control |
| topic | Cardiovascular monitoring generative adversarial networks seismocardiography synthetic generation wearable monitoring |
| url | https://ieeexplore.ieee.org/document/10731564/ |
| work_keys_str_mv | AT jamesskoric generationofseismocardiographyheartbeatsusingawassersteingenerativeadversarialnetworkwithfeaturecontrol AT yannickdmello generationofseismocardiographyheartbeatsusingawassersteingenerativeadversarialnetworkwithfeaturecontrol AT davidvplant generationofseismocardiographyheartbeatsusingawassersteingenerativeadversarialnetworkwithfeaturecontrol |