Medical video encryption using novel 2D Cosine-Sine map and dynamic DNA coding.

Abstract

Modern healthcare systems contain a large amount of sensitive information related to a patient in textual and visual form. Surgical videos and diagnosis data such as ultrasound, Computed Tomography (CT) scan, and Magnetic Resonance Imaging (MRI) are examples of healthcare video data. The secure storage and transmission of medical data have become an important issue in medical applications. To handle this challenge, chaos based cryptosystems are widely used these days. The work in this paper proposes a novel 2D Cosine-Sine map that exploits the existing Sine map and cosine transformation in its mathematical computation. The performance assessment of the suggested map indicates that it possesses a broader chaotic range, more dynamic and hyperchaotic nature when compared to existing chaotic maps. The proposed work, also, combines this novel 2D Cosine-Sine map with dynamic DNA encoding to propose a new scheme to encrypt medical videos. The approach consists primarily of four distinct phases. In the initial stage, the video is divided into frames, and for each frame, a chaotic sequence is generated using a 2D Cosine-sine map. During the second stage, we proceed to pixel permutation for each frame by utilizing the chaotic sequence. The third step is diffusion, in this, we integrate the 2D Cosine-Sine map with dynamic DNA encoding and apply double DNA operations to substitute the pixel values. The last step is DNA decoding to get the frames back in binary format. DNA encoding/decoding rules and operands of DNA operations are not fixed. They are selected dynamically using a random key for each frame. The dynamic selection of encoding/decoding rules and operands is a unique feature that enhances the scheme's security. Simulation results and security analysis proves that the proposed medical video encryption scheme can resist different types of attacks.