Cardiac Abnormality Detection in 12-lead ECGs with Deep Convolutional Neural Networks Using Data Augmentation
Weber L, Gaiduk M, Scherz WD, Seepold R (2020)
Publication Type: Conference contribution
Publication year: 2020
Publisher: IEEE Computer Society
Book Volume: 2020-September
Conference Proceedings Title: Computing in Cardiology
Event location: Rimini, ITA
ISBN: 9781728173825
Abstract
A residual neural network was adapted and applied to the Physionet/Computing data in Cardiology Challenge 2020 to detect 24 different classes of cardiac abnormalities from 12-lead. Additive Gaussian noise, signal shifting, and the classification of signal sections of different lengths were applied to prevent the network from overfitting and facilitating generalization. Due to the use of a global pooling layer after the feature extractor, the network is independent of the signal's length. On the hidden test set of the challenge, the model achieved a validation score of 0.656 and a full test score of 0.27, placing us 15th out of 41 officially ranked teams (Team name: UC_Lab_Kn). These results show the potential of deep neural networks for application to raw data and a complex multi-class multi-label classification problem, even if the training data is from diverse datasets and of differing lengths.
Authors with CRIS profile
Involved external institutions
Germany (DE)How to cite
APA:
Weber, L., Gaiduk, M., Scherz, W.D., & Seepold, R. (2020). Cardiac Abnormality Detection in 12-lead ECGs with Deep Convolutional Neural Networks Using Data Augmentation. In Computing in Cardiology. Rimini, ITA: IEEE Computer Society.
MLA:
Weber, Lucas, et al. "Cardiac Abnormality Detection in 12-lead ECGs with Deep Convolutional Neural Networks Using Data Augmentation." Proceedings of the 2020 Computing in Cardiology, CinC 2020, Rimini, ITA IEEE Computer Society, 2020.
BibTeX: Download