SECURE MEDICAL IMAGE TRANSMISSION USING REVERSIBLE NON-SEPARABLE ENCRYPTION AND OPTIMAL DATA HIDING

Abstract

Secure transfer and storage of medical images are critical requirements in modern healthcare systems, particularly with the increasing adoption of telemedicine, cloud-based diagnostics, and electronic health archives. Traditional encryption techniques such as AES, chaos-based methods, and LSB-based hiding face major limitations including low embedding capability, high computation cost, and reduced protection against differential attacks. To overcome these challenges, this study proposes a reversible and non-separable hybrid medical image encryption scheme integrated with optimal data hiding to ensure both security and high visual fidelity. The proposed model effectively leverages advanced cryptographic mechanisms to enhance confidentiality, integrity, and reversibility while supporting real-time deployment. Experimental results show significant improvements across key performance metrics, achieving over 60 dB PSNR, SSIM ≈ 1, NPCR 99.64%, UACI 33.45%, and 25–30% data embedding capacity, with reduced encryption and decryption latency (5.4 ms and 5.2 ms, respectively). Comparative evaluation against state-of-the-art techniques including AES, chaotic encryption, and hybrid DNA–cryptography confirms superior resilience to statistical and differential attacks with minimal computational overhead. The proposed encryption scheme thus offers a robust, efficient, and practical security solution for safeguarding sensitive medical data in big-data-driven healthcare environments.