CSMI-AW: Computational System for Medical Image Authentication Using Watermarking

Abstract

The enormous growth of digital data in the era of information technology (IT) has provided broader scope towards improving the quality of life. However, the technological aspects in various data-driven solutions also resulted in data duplication and malevolent manipulation during the digital data distributions through networks, especially when it comes to images. Thereby the demand to protect the privacy of image data is highly envisioned and a crucial concern in the current scenario. Currently, medical facilities also require protecting medical image data ownership with copyright protection and content authentication to preserve the authenticity of the patient clinical details. The study in this regard introduces a computationally efficient technique, namely-computational system for medical image authentication using watermarking (CSMI-AW) to protect the ownership of the patient chest x-ray images. The system uses the potential factors associated with the spectral domain techniques, such as with Discrete Cosine Transformation (DCT) approach with non-blind detection. The optimized procedure of DCT-based significant technique for message image/watermarking and substitution methods for embedding process makes the CSMI-AW more robust to retain the medical image watermark and watermarked image perceptual quality. It also simplifies the algorithm steps in such a way where it optimizes the performance of CSMI-AW with a proper balance between robustness, computational cost, and embedding efficiency. However, the CSMI-AW system not only ensures a higher degree of security for digital x-ray images but also attains better computational performance with higher perceptual quality of the watermark. It attains approximately 40% better outcome of PSNR and 60% computational improvement. The comparative performance analysis for the traditional approaches shows its effectiveness in terms of computational efficiency and peak-signal-to-noise ratio (PSNR).