Fast Image Encryption Framework for Medical Images

Abstract

The role of Information and Communication Technology (ICT) in healthcare has increased tremendously. The era of Artificial Intelligence, the Internet of Health Things (IoHT), and networked technologies have revolutionized the paradigm of healthcare. In the healthcare domain, digital images are a means by which sensitive information about the patients can be transferred wirelessly. Maintaining the confidentiality of these images requires designing a cryptosystem that is fast and sturdy. In this work, we present a fast chaos-based encryption scheme for medical images. The scheme employs a Logistic map, Chebyshev map, and Piecewise Linear Chaotic Map (PWLCM) for confusion and diffusion of the medical image. Initially, the image is circularly shifted, and subsequently, bit plane slicing is performed. The Most Significant Bit (MSB) plane is replaced with a plane formed by the XOR operation between the MSB and 7th ISB plane. The resultant image is scrambled using the Pseudo Random Number (PRN) generated by the Logistic map. The scheme is adaptive and calculates the image parameter like sum or mean to determine the initial condition for the PWLCM chaotic map. The PWLCM is used to generate a key image that is XORed with the scrambled image. Finally, a Chebyshev map is iterated and a PRN sequence is generated to permute the image pixels and obtain the final encrypted image. The cryptosystem has been evaluated for multiple types and sizes of medical images and on natural images as well. The average encryption time for an image of size 256×256 is 0.3287 seconds. The entropy is more than 99.70% and the correlation coefficient is close to zero for all the images. Simulation results validate that the cryptosystem has comparable performance with state-of-the-art schemes.