Reversible Data Embedding using Color DICOM Images for Medical Applications

Abstract

Nowadays, the transmission of digitized medical records has become convenient due to widespread knowledge of Internet. The digital technology permits us to apply various steganographic techniques to hide information in digital images. Steganography is an ancient art of hiding data in unsuspected objects like text, image, audio & video whose presence cannot be detected. The copyright protection of the medical records should be ensured as it is transmitted over the Internet. In prior, data hiding techniques that were used to conceal patient’s records into medical images were reversible for grayscale. This paper proposes a reversible data hiding technique along with simultaneous storage of patient information with color medical images. The original image can be completely recovered from the stego-image after the embedded data has been extracted providing better accuracy in diagnosis. It utilizes a difference expansion for quads method to embed the bitstream in the Least Significant Bits. The security is enhanced by selecting the quad sequence which emphasizes a random walk sequence along with intra plane data embedding to embed data across color components giving higher embedding capacity along with good visual quality having PSNR values greater than 37.