Medical image encryption based on Josephus scrambling and dynamic cross-diffusion for patient privacy security

In recent years, smart healthcare mode has been gradually maturing with the evolution of Internet cloud computing technology. However, due to the open and shared features of the Internet, there are a substantial amount of concerns about how to protect patient privacy security in the smart healthcare mode. Hence, this paper proposes a medical image encryption algorithm using the Josephus scrambling and dynamic cross-diffusion techniques for protecting patient privacy. Firstly, we design a novel hyperchaotic system with a broader chaotic interval that can generate chaotic sequences with strong pseudo-randomness. Then, a new dynamic Josephus scrambling methodology based on chaotic sequences is proposed. The scrambling scheme can effectively scramble the image pixel positions, thereby damaging the correlation between adjacent pixels of the target image. Finally, we devise a dynamic parallel cross-diffusion scheme combined with chaotic sequences to further encrypt the scrambled image. Meanwhile, the SHA-256 combined with the target image to generate the hash value is utilized for updating the initial key of the algorithm. Simulation experiments reveal that the devised encryption algorithm can convert the target image into a meaningless snowflake image. And security analysis results show that the presented algorithm is robust to common cryptanalysis approaches. In addition, we compare the corresponding performance indicators with those of some state-of-the-art image algorithms, and the results demonstrate that our algorithm can exhibit superior encryption performance.