Secure Exchange of Medical Images Via Extended Visual Cryptography

Abstract

Medical image security is acquiring its importance to preserve the integrity and confidentiality of information (medical data) from malicious users given its importance in timely and successful diagnosis. In this context, several techniques have been developed to protect medical images, such as encryption, data hiding, image tagging, application of Hash algorithms, etc. This paper proposes a technique to cipher medical images by adding the metadata inside a cover image, based on extended visual cryptography as well as the inclusion of a Hash-like function to verify the integrity of the image and the metadata once they are recovered. The method proposed in this work is implemented using medical images with a grayscale resolution of [0,4095] that is a depth of 12 bits/pixel and color images with 24 bits/pixel depth. Experimental results prove the effectiveness of the proposed method in the task of secure exchange of medical images by allowing higher hiding capability, lower distortion in the visual quality of the image with the hidden medical data, as well as a means to verify the integrity of the sent data, compared to state-of-the-art.