EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Improving Adversarial Robustness of ECG Classification Based on Lipschitz Constraints and Channel Activation Suppression

Xin Chen, Yujuan Si, Zhanyuan Zhang, Wenke Yang, Jianchao Feng

Sensors · 06.05.2024 · 24 (9) · 2954

DOI: 10.3390/s24092954 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 06.05.2024. Verfasst von Xin Chen und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Electronic and Information Engineering (SEIE), Zhuhai College of Science and Technology, Zhuhai 519041, China. Hauptthemen laut MEDLINE: Electrocardiography, Neural Networks, Computer, Algorithms. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the Natural Science Foundation of Guangdong Province, the Guangdong Key Disciplines Project, the Featured Innovation Projects of the Guangdong Universities.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Deep neural networks (DNNs) are increasingly important in the medical diagnosis of electrocardiogram (ECG) signals. However, research has shown that DNNs are highly vulnerable to adversarial examples, which can be created by carefully crafted perturbations. This vulnerability can lead to potential medical accidents. This poses new challenges for the application of DNNs in the medical diagnosis of ECG signals. This paper proposes a novel network Channel Activation Suppression with Lipschitz Constraints Net (CASLCNet), which employs the Channel-wise Activation Suppressing (CAS) strategy to dynamically adjust the contribution of different channels to the class prediction and uses the 1-Lipschitz’s ℓ∞ distance network as a robust classifier to reduce the impact of adversarial perturbations on the model itself in order to increase the adversarial robustness of the model. The experimental results demonstrate that CASLCNet achieves ACCrobust scores of 91.03% and 83.01% when subjected to PGD attacks on the MIT-BIH and CPSC2018 datasets, respectively, which proves that the proposed method in this paper enhances the model’s adversarial robustness while maintaining a high accuracy rate.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

ElectrocardiographyNeural Networks, ComputerAlgorithmsHumansSignal Processing, Computer-Assisted

Stichwörter der Autor:innen

adversarial robustnessarrhythmia classificationchannel-wise activation suppressingℓ∞ distance network

Autor:innen und Institutionen

  • School of Electronic and Information Engineering (SEIE), Zhuhai College of Science and Technology, Zhuhai 519041, China.
  • College of Communication Engineering, Jilin University, Changchun 130012, China.

Förderung

  • the Natural Science Foundation of Guangdong Province
  • the Guangdong Key Disciplines Project
  • the Featured Innovation Projects of the Guangdong Universities
  • Natural Science Foundation of Guangdong Province
  • Guangdong Key Disciplines Project
  • Featured Innovation Projects of the Guangdong Universities

Zitieren

APA

Chen, X., Si, Y., Zhang, Z., Yang, W., & Feng, J. (2024). Improving Adversarial Robustness of ECG Classification Based on Lipschitz Constraints and Channel Activation Suppression. Sensors, 24(9), 2954. https://doi.org/10.3390/s24092954

Vancouver

Chen X, Si Y, Zhang Z, Yang W, Feng J. Improving Adversarial Robustness of ECG Classification Based on Lipschitz Constraints and Channel Activation Suppression. Sensors. 2024;24(9):2954. doi:10.3390/s24092954

BibTeX
@article{Chen2024,
  author = {Chen, Xin and Si, Yujuan and Zhang, Zhanyuan and Yang, Wenke and Feng, Jianchao},
  title = {Improving Adversarial Robustness of ECG Classification Based on Lipschitz Constraints and Channel Activation Suppression},
  journal = {Sensors},
  year = {2024},
  volume = {24},
  number = {9},
  pages = {2954},
  doi = {10.3390/s24092954}
}
Kostenlos starten →