Hardware-Accelerated Least Significant Bit Framework: A Low Cost Approach to Securing Clinical Data

Information Technology is continuously making communication much easier thereby drawing the interest of millions of users, including medics, to data communications. However, security of the information communicated still remains issues of concern, especially when it comes to patients’ confidential information. Apparently, various concepts have being proposed to solving most of the problems of information security, such as the cryptography and the steganography. Image steganography approach focuses on concealing the existence of information from unintended person. Least Significant Bit (LSB) approach of hiding data is one of the most popular in this regards but with weakness of low embedding capacity and therefore low security. Consequently, there have been several attempts by researchers to improving the embedding capacity and low security in LSB. This study therefore proposes hardware-accelerated steganography data hiding framework towards addressing the weaknesses of low security and embedding capacity exhibited in LSB data hiding approach. Within Compute Unified Device Architecture (CUDA), the proposed framework was able to conceal secret data using the LSB approach where the pixels of the cover image and the secret data were converted into their equivalent binary bits, and the binary bits grouped into bytes of 8-bits each. The LSB of the cover image was replaced with the bits of the secret data starting with the most significant bit of each byte of 8-bit of the secret data, until the whole data is fully embedded in the cover image to generate the stego-image. The proposed framework was further evaluated for possible security of the embedded secret data by encrypting the stego-image with password. The experimental results show that the images in PNG, JPG and JPEG format tend to embed and hide the secret data successfully but with increase in the size of the generated stego-image, while images in TIFF and BMP format tend to embed and hide the secret data successfully and retain their original sizes. Interestingly, the entire data hiding stages were achieved within 3s for up to 110KB of data and not more than 4s for 192KB.