An Efficient Medical Image Cryptosystem Based on Chaotic Maps

Abstract

With the growing applications of digital images in medical health system and the development of computer networks, medical image encryption has become an important technology to protect patients' privacy when it is transmitted over open networks. Recently, chaos-based encryption approaches have been proven to be more effective for image encryption than other traditional methods (e.g. IDEA and AES). In this paper, we propose a novel chaos-based medical image cryptosystem with permutationdiffusion architecture, which provides efficient and secure transmissions for medical images over open networks. In the permutation stage, we introduce a novel shuffling method, which shuffles each pixel in plain image by exchanging it with a certain pixel selected by chaotic Logistic map. Compared with conventional permutation schemes based on Arnold cat map, Baker map and Standard map, the new scheme is superior in both key space and efficiency. In diffusion stage, each pixel is masked according to both key stream element and previous ciphered pixel so as to make the cryptosystem secure against differential attack. Intensive cryptanalysis is carried out and the experimental results demonstrate that the cryptosystem can withstand all kinds of known attacks, which includes brute-force attack, differential attack, known/chosen plain-text attack as well as various statistical attacks. Moreover, efficiency test indicates that the cryptosystem is efficient enough for real-time medical image protection applications.