Analysis and Computation of Encryption Technique to Enhance Security of Medical Images

Recently, there are many advanced techniques used in the medical fields, such as smart health, e-health and telemedicine applications. These techniques rely on open source networks to transmit the digital medical images. These images have very sensitive and confidential information about the patients. Regrettably, most of the regularly used algorithms provide very less security and present a high communication overhead, with high computation costs. Due to these issues, a Region of Interest (ROI) method, based on selective image encryption, is proposed in this paper. An active contour image segmentation is first used to divide the ROI and the Region of Background (ROB). Then, a Hilbert curve, with a Skew Tent map are used to implement the Permutation and diffusion techniques. Namely, the ROI part is permuted according to a Hilbert scan pattern, to decrease the similarities between the inside adjacent pixels. Then pixels in the permuted ROI part are XOR pixel-wise with random numbers, generated based on a predefined threshold value from a Skew Tent map. Finally, encrypted ROI and ROB blocks are combined to generate encrypted images. Given the obtained experimental results, the proposed method is very likely to improve image security, through the employment of correlation checking, key sensitivity, entropy, diffusion characteristic and histogram tools.