Hiding health report in X-ray images to protect people privacy

Abstract

X-rays images are a popular medical diagnoses method. These images are hard to read and to be understandable outside the medical community. However, the radiologist writes a report of the X-ray. This report can be understandable easily which, can impact the privacy of the patients’ medical records. To tackle this issue, in this work, X-ray image steganography algorithm based on Sudoku mathematical game is proposed. The method converts the image into triangular shaped blocks. Subsequently, the data is encoded and segmented two bits at the time to be embedded in the image. To evaluate the algorithm, eight different X-ray images from all X-ray types have been utilized for data hiding process. The algorithm has been compared to the lest significant bit (LSB) and the turtle shell algorithms using the mean square error (MSE) and peak signal to noise ratio (PSNR) performance metric. The proposed algorithm obtained higher bit per pixel embedding capacity with 2 bpp which is higher than LSB and turtle shell. Moreover, the algorithm has higher PSNR compared to turtle shell algorithm.