DeepFake electrocardiograms using generative adversarial networks are the beginning of the end for privacy issues in medicine
Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt
Standard
DeepFake electrocardiograms using generative adversarial networks are the beginning of the end for privacy issues in medicine. / Thambawita, Vajira; Isaksen, Jonas L; Hicks, Steven A; Ghouse, Jonas; Ahlberg, Gustav; Linneberg, Allan; Grarup, Niels; Ellervik, Christina; Olesen, Morten Salling; Hansen, Torben; Graff, Claus; Holstein-Rathlou, Niels-Henrik; Strümke, Inga; Hammer, Hugo L.; Maleckar, Mary M.; Halvorsen, Pål; Riegler, Michael A; Kanters, Jørgen K.
I: Scientific Reports, Bind 11, Nr. 1, 21896, 2021.Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - JOUR
T1 - DeepFake electrocardiograms using generative adversarial networks are the beginning of the end for privacy issues in medicine
AU - Thambawita, Vajira
AU - Isaksen, Jonas L
AU - Hicks, Steven A
AU - Ghouse, Jonas
AU - Ahlberg, Gustav
AU - Linneberg, Allan
AU - Grarup, Niels
AU - Ellervik, Christina
AU - Olesen, Morten Salling
AU - Hansen, Torben
AU - Graff, Claus
AU - Holstein-Rathlou, Niels-Henrik
AU - Strümke, Inga
AU - Hammer, Hugo L.
AU - Maleckar, Mary M.
AU - Halvorsen, Pål
AU - Riegler, Michael A
AU - Kanters, Jørgen K
N1 - © 2021. The Author(s).
PY - 2021
Y1 - 2021
N2 - 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.
AB - 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.
U2 - 10.1038/s41598-021-01295-2
DO - 10.1038/s41598-021-01295-2
M3 - Journal article
C2 - 34753975
VL - 11
JO - Scientific Reports
JF - Scientific Reports
SN - 2045-2322
IS - 1
M1 - 21896
ER -
ID: 284634905