Fully Reversible Steganography with Authentication in Wavelet Domain for Telemedicine Applications

Abstract

Last year has witnessed a growing reliance on telemedicine practices requiring the transmission of medical images along with sensitive patient records and diagnosis information to/from remote locations. However, maintaining optimal security and authentication of appended information along with complete reversibility of medical image is a challenging telemedicine task. Traditional approaches to medical image steganography attempt to hide a large amount of information and make it imperceptible at the cost of full reversibility of all regions in the medical image. In this paper, we propose an Integer Wavelet Transform (IWT) based efficient approach to strike an optimal balance between large hiding capacity and imperceptibility while attaining full reversibility of the entire underlying medical image. The redundancy present in multiple optimally selected bit-planes of the detail coefficients of IWT is exploited to hide a large amount of data along with encoded transform bit-plane original data for full reversibility. For authentication purposes, the hospital logo is also embedded in the image for complete recovery and verification at the receiver end. The experimental results demonstrated on CT, MRI, and PET datasets, show improved steganography performance in terms of capacity without appreciable loss in imperceptibility and without compromising full reversibility of the entire medical image.