Hiding patient information in medical images:A high-capacity and reversible hiding algorithm for E-healthcare
Xiaoyi Zhou, Shuai Lee · 2023
The rise of online medical-related technologies has brought great convenience to doctors and patients. However, electronic patient information may be subject to attacks and tampering during medical transmission, leading to compromised information. Therefore, it is necessary to protect the security of data in medical transmission by authenticated electronic patient information. This paper proposes a high-capacity reversible data hiding technology based on interpolated images, which can successfully protect patients’ confidential information and verify received content. By using predictors to generate high-quality cover images, and then using data foding strategy to hide electronic patient information into medical images, the entire scheme can be divided into image preprocessing, data processing, and data embedding processes. Image preprocessing includes generating predicted images and performing boundary value on them to generate high-quality covered images. Data processing encrypts the secret data, divides the encrypted data into n blocks of 3 bits, and converts each block into a decimal number. Before embedding the secret data, these numbers are subjected to data foding strategy operation, and data labeling is carried out to ensure the recovery of secret data. The embedding process divides the cover image into 2x2 blocks and embeds secret data in the non-seed pixel diagonal of each block. Last but not least, a fragile watermark is also embedded in the cover image to determine whether the transmitted information has been tampered with. Experimental results show that the method can achieve a high embedding capacity of an average of 2.25 bits per pixel, while the average peak signal-to-noise ratio of the steganographic image reaches 42dB, which is good compared to existing technologies.