A Secure Zero-Watermarking Framework for 3D Medical Data Using Hyper chaos and Dual-Tree Complex Wavelet Transform in IoMT

Abstract

Recent technological breakthroughs in 5g, higher data, and cloud primarily affecting IoT, specifically the IoMT have paved a way of generating and analysing large 3D volumetric medical data for enhancing diagnostics and pathological analysis. However, the increasingly adoption of IoMT has increased risks associated with personal data protection, including data privacy. To design these solutions, a new reliable zero-watermarking algorithm based on 3D hyper chaos and 3D dual-tree complex wavelet transform (DTCWT) is presented. In the proposed approach, the human visual features will be combined with perceptual hashing techniques to produce a robust binary sequence. The watermark is first encrypted with 3D hyper chaos cryptographic system for enhancing the security; the 3D DTCWT-DCT transforms are used to obtain the low frequency coefficients for derivation of a secure watermark key. This zero-embedding framework guarantees the accuracy of the medical volume data meeting the strict need for diagnostic precision as well as data credibility.   Accumulated simulation results confirm that the designed algorithm is efficient. With an average of 98.35% Normalized Correlation (NC) under standard conditions, the proposed method outperforms state of the art in terms of NC by 46.67% under geometric attacks. Moreover, for the enhancement of the image, the proposed algorithm outperforms existing methods in terms of Peak Signal to Noise Ratio (PSNR): 52.14 dB and Structural Similarity Index (MSSIM) of 0.984 shows better results, less distortion and more robustness. The two results imply that the algorithm effectively guarantees safe, efficient, and accurate data exchange and storage in IoMT processes.