Stego-eHealth: An eHealth System for Secured Transfer of Medical Images using Image Steganography

Abstract

COVID pandemic has necessitated the need for virtual and online health care systems to avoid contacts. The transfer of sensitive medical information including the chest and lung X-ray happens through untrusted channels making it prone to many possible attacks. This paper aims to secure the medical data of the patients using image steganography when transferring through untrusted channels. A deep learning method with three parts is proposed – preprocessing module, embedding network and the extraction network. Features from the cover image and the secret image are extracted by the preprocessing module. The merged features from the preprocessing module are used to output the stego image by the embedding network. The stego image is given as the input to the extraction network to extract the ingrained secret image. Mean Squared Error (MSE) and Peak Signal-to-Noise Ratio (PSNR) are the evaluation metrics used. Higher PSNR value proves the higher security; robustness of the method and the image results show the higher imperceptibility. The hiding capacity of the proposed method is 100% since the cover image and the secret image are of the same size.