Implementation of Secure Lossless Compression of Medical Images Based on Multi- Model Frame Work

Abstract

The increasing volume of medical imaging data demands efficient compression techniques without compromising image quality. Security is also critical for protecting sensitive patient data. This study aimed at develop a Secure lossless compression of medical images based on multi model frame work and RSA encryption for secure medical image storage and transmission. The framework was implemented using Python, with datasets obtained from Kaggle Medical Imaging Database. NTC-MMF was used for compression, while hybrid RSAAES encryption was applied for security. The system was evaluated based on compression ratio, processing time, image quality (SSIM, PSNR), and encryption speed. The proposed system achieved an average compression ratio of 1.6 for PNG and 3.2 for JPG, with PNG retaining full image quality (SSIM: 1.0, PSNR: ∞ dB). Compression time was lower for JPG (~500 ms) than PNG (~1200 ms). Encryption time was ~1000 ms for PNG and ~600 ms for JPG, proving that file size impacts security processing speeds. The study successfully developed an optimized compression and security framework that balances image fidelity, storage efficiency, and data protection. These findings can enhance medical image management in hospitals and cloud-based health systems.