EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Secure and Intelligent Cardiovascular Risk Prediction with Homomorphic Encryption and Hybrid GWO-ABO Optimized CNN-BiLSTM

Farhana Begum, Y. Md Riyazuddin

Engineering, Technology & Applied Science Research · 11.08.2026 · 16 (4) · 37347-37353

DOI: 10.48084/etasr.17946 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Engineering, Technology & Applied Science Research, online erschienen am 11.08.2026. Verfasst von Farhana Begum und 1 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In this study, a hybrid Convolutional Neural Network-Bidirectional Long Short-Term Memory (CNN-BiLSTM) model is proposed for heart disease risk prediction using Electrocardiography (ECG) data, while ensuring the privacy and security of sensitive patient data. For the training and evaluation of the proposed model, the publicly accessible Physikalisch-Technische Bundesanstalt Extended Large (PTB-XL) ECG dataset, containing over 21,000 annotated ECG recordings, was used. ECG signals underwent preprocessing and normalization techniques to eliminate noise and prepare the data for deep learning analysis. To improve model performance, the hybrid Grey Wolf Optimization-Artificial Butterfly Optimization (GWO-ABO) algorithm was employed to automatically tune key hyperparameters, including dropout rate, learning rate, and filter size, while Homomorphic Encryption (HE) was used to encrypt medical data and maintain patient privacy. The model achieved an accuracy of 84.06%, a precision of 82.93%, a recall of 89.47%, and an F1-score of 86.08% on the test set, while it maintained a good balance between prediction accuracy and data privacy, making it suitable for secure implementation in real-time clinical decision support and cloud-based healthcare system applications. Future research can combine federated learning with encryption techniques to further improve privacy-preserving management of sensitive medical data.

Zitieren

APA

Begum, F., & Riyazuddin, Y. M. (2026). Secure and Intelligent Cardiovascular Risk Prediction with Homomorphic Encryption and Hybrid GWO-ABO Optimized CNN-BiLSTM. Engineering, Technology & Applied Science Research, 16(4), 37347-37353. https://doi.org/10.48084/etasr.17946

Vancouver

Begum F, Riyazuddin YM. Secure and Intelligent Cardiovascular Risk Prediction with Homomorphic Encryption and Hybrid GWO-ABO Optimized CNN-BiLSTM. Engineering, Technology & Applied Science Research. 2026;16(4):37347-37353. doi:10.48084/etasr.17946

BibTeX
@article{Begum2026,
  author = {Begum, Farhana and Riyazuddin, Y. Md},
  title = {Secure and Intelligent Cardiovascular Risk Prediction with Homomorphic Encryption and Hybrid GWO-ABO Optimized CNN-BiLSTM},
  journal = {Engineering, Technology & Applied Science Research},
  year = {2026},
  volume = {16},
  number = {4},
  pages = {37347--37353},
  doi = {10.48084/etasr.17946}
}
Kostenlos starten →