International Journal of Computer Techniques Volume 12 Issue 5 | Next-Generation ECG Steganography and Prediction through Deep Learning Techniques
Next-Generation ECG Steganography and Prediction through Deep Learning Techniques
Author: Soumyendu Banerjee
Department of Electrical Engineering, Institute of Engineering and Management, University of Engineering and Management, Kolkata, India
Email: banerjeesoumyendu@gmail.com
Journal: International Journal of Computer Techniques (IJCT)
Volume: 12 | Issue: 5 | Publication Date: September – October 2025
ISSN: 2394-2231 | Page: 30 | Journal URL: https://ijctjournal.org/
Abstract
This paper introduces a secure and efficient ECG steganography framework using deep learning techniques. Patient data is embedded in TP-segments to preserve diagnostic integrity, while LSTM networks reconstruct signals post-extraction. The method achieves high fidelity and robustness, outperforming traditional frequency-domain approaches. Experimental results on MIT-BIH, PTB, and European ST-T datasets confirm its suitability for telecardiology and secure biomedical communication.
Keywords
ECG Steganography, Deep Learning, LSTM Recurrent Neural Network, Biomedical Signal Security, TP-Segment Prediction
Conclusion
The proposed framework ensures secure ECG data transmission while maintaining clinical accuracy. Deep learning integration enhances signal recovery and imperceptibility. Future work may include multi-lead ECG testing, layered encryption, and real-world deployment in telemedicine systems.
References
Includes 29 references from IEEE, Elsevier, Springer, and other peer-reviewed sources covering ECG steganography, signal compression, LSTM prediction, and biomedical encryption protocols.
