A Low-Degradation Steganography Model for Data Hiding in Medical Images

P. Mortazavian, M. Jahangiri, and E. Fatemizadeh (Iran)


Medical images, Image Steganography, data hiding.


Textual labels overlaid on medical images provide significant information which should be unavailable to un-authorized persons who have access to the image. This paper proposes "image steganography" as a means to hide this valuable information inside the image without losing it. Conventional template matching is used to find the text on the image and extract it, the text is subsequently removed from the image and In the next step, a steganographical method is proposed by the authors to embed the text into the image, so that the textual data will be present in the image but visually undetectable. In the proposed steganographical scheme, the data is embedded into the mean value of a number of pixels. But since any degradation in medical images could reduce the clinical value of the image, keeping the image alterations to a negligible amount becomes of great concern , thus the maximum modification on each pixel is limited by the algorithm. Implementation of this algorithm on many medical images shows that a reasonable amount of data can be embedded into an image without perceptible alterations besides having a satisfying robustness against noise.

Important Links:

Go Back