Inherently explainable deep neural network-based interpretation of electrocardiograms using variational auto-encoders
View ORCID ProfileRutger R. van de Leur, Max N. Bos, Karim Taha, Arjan Sammani, Stefan van Duijvenboden, Pier D. Lambiase, Rutger J. Hassink, Pim van der Harst, Pieter A. Doevendans, Deepak K. Gupta, René van Es
doi: https://doi.org/10.1101/2022.01.04.22268759
Rutger R. van de Leur
aDepartment of Cardiology, University Medical Center Utrecht, Utrecht, The Netherlands
bNetherlands Heart Institute, Utrecht, The Netherlands
MDMax N. Bos
aDepartment of Cardiology, University Medical Center Utrecht, Utrecht, The Netherlands
cInformatics Institute, University of Amsterdam, Amsterdam, The Netherlands
MScKarim Taha
aDepartment of Cardiology, University Medical Center Utrecht, Utrecht, The Netherlands
bNetherlands Heart Institute, Utrecht, The Netherlands
MDArjan Sammani
aDepartment of Cardiology, University Medical Center Utrecht, Utrecht, The Netherlands
MDStefan van Duijvenboden
dInstitute of Cardiovascular Science, University College London, London, United Kingdom
PhDPier D. Lambiase
dInstitute of Cardiovascular Science, University College London, London, United Kingdom
MD PhDRutger J. Hassink
aDepartment of Cardiology, University Medical Center Utrecht, Utrecht, The Netherlands
MD PhDPim van der Harst
aDepartment of Cardiology, University Medical Center Utrecht, Utrecht, The Netherlands
MD PhDPieter A. Doevendans
aDepartment of Cardiology, University Medical Center Utrecht, Utrecht, The Netherlands
bNetherlands Heart Institute, Utrecht, The Netherlands
eCentral Military Hospital, Utrecht, The Netherlands
MD PhDDeepak K. Gupta
cInformatics Institute, University of Amsterdam, Amsterdam, The Netherlands
PhDRené van Es
aDepartment of Cardiology, University Medical Center Utrecht, Utrecht, The Netherlands
PhDData Availability
The training datasets used in this study are not openly available due to privacy concerns. The expert-annotated test set is available upon request to the corresponding author. The decoder for the FactorECG is publicly available at https://decoder.ecgx.ai. The code for training and evaluating the β-VAE and the black box DNN is available upon request to the corresponding author.
Posted January 05, 2022.
Inherently explainable deep neural network-based interpretation of electrocardiograms using variational auto-encoders
Rutger R. van de Leur, Max N. Bos, Karim Taha, Arjan Sammani, Stefan van Duijvenboden, Pier D. Lambiase, Rutger J. Hassink, Pim van der Harst, Pieter A. Doevendans, Deepak K. Gupta, René van Es
medRxiv 2022.01.04.22268759; doi: https://doi.org/10.1101/2022.01.04.22268759
Inherently explainable deep neural network-based interpretation of electrocardiograms using variational auto-encoders
Rutger R. van de Leur, Max N. Bos, Karim Taha, Arjan Sammani, Stefan van Duijvenboden, Pier D. Lambiase, Rutger J. Hassink, Pim van der Harst, Pieter A. Doevendans, Deepak K. Gupta, René van Es
medRxiv 2022.01.04.22268759; doi: https://doi.org/10.1101/2022.01.04.22268759
Subject Area
Subject Areas
- Addiction Medicine (380)
- Allergy and Immunology (697)
- Anesthesia (187)
- Cardiovascular Medicine (2824)
- Dermatology (242)
- Emergency Medicine (427)
- Epidemiology (12522)
- Forensic Medicine (10)
- Gastroenterology (799)
- Genetic and Genomic Medicine (4402)
- Geriatric Medicine (399)
- Health Economics (712)
- Health Informatics (2835)
- Health Policy (1044)
- Hematology (372)
- HIV/AIDS (893)
- Medical Education (412)
- Medical Ethics (114)
- Nephrology (460)
- Neurology (4161)
- Nursing (220)
- Nutrition (615)
- Oncology (2189)
- Ophthalmology (623)
- Orthopedics (254)
- Otolaryngology (316)
- Pain Medicine (263)
- Palliative Medicine (81)
- Pathology (484)
- Pediatrics (1169)
- Primary Care Research (481)
- Public and Global Health (6739)
- Radiology and Imaging (1483)
- Respiratory Medicine (896)
- Rheumatology (429)
- Sports Medicine (362)
- Surgery (470)
- Toxicology (57)
- Transplantation (198)
- Urology (173)