DeepFake electrocardiograms using generative adversarial networks are the beginning of the end for privacy issues in medicine
Abstract Recent global developments underscore the prominent role big data have in modern medical science. But privacy issues constitute a prevalent problem for collecting and sharing data between researchers. However, synthetic data generated to represent real data carrying similar information and...
Guardado en:
| Autores principales: | , , , , , , , , , , , , , , , , , |
|---|---|
| Formato: | article |
| Lenguaje: | EN |
| Publicado: |
Nature Portfolio
2021
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/083a56ecd10843639b0fa258ef14862d |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| id |
oai:doaj.org-article:083a56ecd10843639b0fa258ef14862d |
|---|---|
| record_format |
dspace |
| spelling |
oai:doaj.org-article:083a56ecd10843639b0fa258ef14862d2021-11-14T12:21:22ZDeepFake electrocardiograms using generative adversarial networks are the beginning of the end for privacy issues in medicine10.1038/s41598-021-01295-22045-2322https://doaj.org/article/083a56ecd10843639b0fa258ef14862d2021-11-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-01295-2https://doaj.org/toc/2045-2322Abstract Recent global developments underscore the prominent role big data have in modern medical science. But privacy issues constitute a prevalent problem for collecting and sharing data between researchers. However, synthetic data generated to represent real data carrying similar information and distribution may alleviate the privacy issue. In this study, we present generative adversarial networks (GANs) capable of generating realistic synthetic DeepFake 10-s 12-lead electrocardiograms (ECGs). We have developed and compared two methods, named WaveGAN* and Pulse2Pulse. We trained the GANs with 7,233 real normal ECGs to produce 121,977 DeepFake normal ECGs. By verifying the ECGs using a commercial ECG interpretation program (MUSE 12SL, GE Healthcare), we demonstrate that the Pulse2Pulse GAN was superior to the WaveGAN* to produce realistic ECGs. ECG intervals and amplitudes were similar between the DeepFake and real ECGs. Although these synthetic ECGs mimic the dataset used for creation, the ECGs are not linked to any individuals and may thus be used freely. The synthetic dataset will be available as open access for researchers at OSF.io and the DeepFake generator available at the Python Package Index (PyPI) for generating synthetic ECGs. In conclusion, we were able to generate realistic synthetic ECGs using generative adversarial neural networks on normal ECGs from two population studies, thereby addressing the relevant privacy issues in medical datasets.Vajira ThambawitaJonas L. IsaksenSteven A. HicksJonas GhouseGustav AhlbergAllan LinnebergNiels GrarupChristina EllervikMorten Salling OlesenTorben HansenClaus GraffNiels-Henrik Holstein-RathlouInga StrümkeHugo L. HammerMary M. MaleckarPål HalvorsenMichael A. RieglerJørgen K. KantersNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-8 (2021) |
| institution |
DOAJ |
| collection |
DOAJ |
| language |
EN |
| topic |
Medicine R Science Q |
| spellingShingle |
Medicine R Science Q Vajira Thambawita Jonas L. Isaksen Steven A. Hicks Jonas Ghouse Gustav Ahlberg Allan Linneberg Niels Grarup Christina Ellervik Morten Salling Olesen Torben Hansen Claus Graff Niels-Henrik Holstein-Rathlou Inga Strümke Hugo L. Hammer Mary M. Maleckar Pål Halvorsen Michael A. Riegler Jørgen K. Kanters DeepFake electrocardiograms using generative adversarial networks are the beginning of the end for privacy issues in medicine |
| description |
Abstract Recent global developments underscore the prominent role big data have in modern medical science. But privacy issues constitute a prevalent problem for collecting and sharing data between researchers. However, synthetic data generated to represent real data carrying similar information and distribution may alleviate the privacy issue. In this study, we present generative adversarial networks (GANs) capable of generating realistic synthetic DeepFake 10-s 12-lead electrocardiograms (ECGs). We have developed and compared two methods, named WaveGAN* and Pulse2Pulse. We trained the GANs with 7,233 real normal ECGs to produce 121,977 DeepFake normal ECGs. By verifying the ECGs using a commercial ECG interpretation program (MUSE 12SL, GE Healthcare), we demonstrate that the Pulse2Pulse GAN was superior to the WaveGAN* to produce realistic ECGs. ECG intervals and amplitudes were similar between the DeepFake and real ECGs. Although these synthetic ECGs mimic the dataset used for creation, the ECGs are not linked to any individuals and may thus be used freely. The synthetic dataset will be available as open access for researchers at OSF.io and the DeepFake generator available at the Python Package Index (PyPI) for generating synthetic ECGs. In conclusion, we were able to generate realistic synthetic ECGs using generative adversarial neural networks on normal ECGs from two population studies, thereby addressing the relevant privacy issues in medical datasets. |
| format |
article |
| author |
Vajira Thambawita Jonas L. Isaksen Steven A. Hicks Jonas Ghouse Gustav Ahlberg Allan Linneberg Niels Grarup Christina Ellervik Morten Salling Olesen Torben Hansen Claus Graff Niels-Henrik Holstein-Rathlou Inga Strümke Hugo L. Hammer Mary M. Maleckar Pål Halvorsen Michael A. Riegler Jørgen K. Kanters |
| author_facet |
Vajira Thambawita Jonas L. Isaksen Steven A. Hicks Jonas Ghouse Gustav Ahlberg Allan Linneberg Niels Grarup Christina Ellervik Morten Salling Olesen Torben Hansen Claus Graff Niels-Henrik Holstein-Rathlou Inga Strümke Hugo L. Hammer Mary M. Maleckar Pål Halvorsen Michael A. Riegler Jørgen K. Kanters |
| author_sort |
Vajira Thambawita |
| title |
DeepFake electrocardiograms using generative adversarial networks are the beginning of the end for privacy issues in medicine |
| title_short |
DeepFake electrocardiograms using generative adversarial networks are the beginning of the end for privacy issues in medicine |
| title_full |
DeepFake electrocardiograms using generative adversarial networks are the beginning of the end for privacy issues in medicine |
| title_fullStr |
DeepFake electrocardiograms using generative adversarial networks are the beginning of the end for privacy issues in medicine |
| title_full_unstemmed |
DeepFake electrocardiograms using generative adversarial networks are the beginning of the end for privacy issues in medicine |
| title_sort |
deepfake electrocardiograms using generative adversarial networks are the beginning of the end for privacy issues in medicine |
| publisher |
Nature Portfolio |
| publishDate |
2021 |
| url |
https://doaj.org/article/083a56ecd10843639b0fa258ef14862d |
| work_keys_str_mv |
AT vajirathambawita deepfakeelectrocardiogramsusinggenerativeadversarialnetworksarethebeginningoftheendforprivacyissuesinmedicine AT jonaslisaksen deepfakeelectrocardiogramsusinggenerativeadversarialnetworksarethebeginningoftheendforprivacyissuesinmedicine AT stevenahicks deepfakeelectrocardiogramsusinggenerativeadversarialnetworksarethebeginningoftheendforprivacyissuesinmedicine AT jonasghouse deepfakeelectrocardiogramsusinggenerativeadversarialnetworksarethebeginningoftheendforprivacyissuesinmedicine AT gustavahlberg deepfakeelectrocardiogramsusinggenerativeadversarialnetworksarethebeginningoftheendforprivacyissuesinmedicine AT allanlinneberg deepfakeelectrocardiogramsusinggenerativeadversarialnetworksarethebeginningoftheendforprivacyissuesinmedicine AT nielsgrarup deepfakeelectrocardiogramsusinggenerativeadversarialnetworksarethebeginningoftheendforprivacyissuesinmedicine AT christinaellervik deepfakeelectrocardiogramsusinggenerativeadversarialnetworksarethebeginningoftheendforprivacyissuesinmedicine AT mortensallingolesen deepfakeelectrocardiogramsusinggenerativeadversarialnetworksarethebeginningoftheendforprivacyissuesinmedicine AT torbenhansen deepfakeelectrocardiogramsusinggenerativeadversarialnetworksarethebeginningoftheendforprivacyissuesinmedicine AT clausgraff deepfakeelectrocardiogramsusinggenerativeadversarialnetworksarethebeginningoftheendforprivacyissuesinmedicine AT nielshenrikholsteinrathlou deepfakeelectrocardiogramsusinggenerativeadversarialnetworksarethebeginningoftheendforprivacyissuesinmedicine AT ingastrumke deepfakeelectrocardiogramsusinggenerativeadversarialnetworksarethebeginningoftheendforprivacyissuesinmedicine AT hugolhammer deepfakeelectrocardiogramsusinggenerativeadversarialnetworksarethebeginningoftheendforprivacyissuesinmedicine AT marymmaleckar deepfakeelectrocardiogramsusinggenerativeadversarialnetworksarethebeginningoftheendforprivacyissuesinmedicine AT palhalvorsen deepfakeelectrocardiogramsusinggenerativeadversarialnetworksarethebeginningoftheendforprivacyissuesinmedicine AT michaelariegler deepfakeelectrocardiogramsusinggenerativeadversarialnetworksarethebeginningoftheendforprivacyissuesinmedicine AT jørgenkkanters deepfakeelectrocardiogramsusinggenerativeadversarialnetworksarethebeginningoftheendforprivacyissuesinmedicine |
| _version_ |
1718429209950420992 |