Myriad Wavelet Transformed Certificateless Signcryptive Extreme Learning Steganography for Secure Medical Image Transmission

Abstract

Medical imaging is a vital part of the healthcare sector which facilitates the communication of medical images like X-rays, MRIs, and CT scans from one place to another. Medical images are now being sent over public networks due to improvements in the healthcare industry, which creates potential security challenges like authentication, integrity, and confidentiality. The medical images size being transmitted has also become a major concern after the rapid growth of computer networks and information technology.  Therefore, ensuring secure medical image transmission and efficient compression are crucial aspects of modern healthcare systems. To enhance the security of image transmission while reducing image size, an efficient technique known as Myriad Wavelet Transformed Certificateless Signcryption Extreme Learning Steganography (MWTCSELS) has been introduced. The MWTCSELS involves four distinct processes namely image preprocessing, image compression, signcryption, and embedding. The first step is to denoise the medical image, the Wilcox indexive myriad filtering technique is used. Then after the preprocessing is done by compressing the image and minimizing the storage space in the communication, the Burrows-Wheeler Hilbert linear curve transform is used. In the third step, the Schmidt-Samoa cryptographic Certificateless Signcryption method is employed to encrypt the input image securely. Lastly, the Mar Wavelet transformed Extreme Learning Machine is used to embed confidential data into the image using a Stego key. The resulting Stego images can be transferred to the receiver end. The original images are restored from the Stego images by the authorized receiver after performing the extraction process. Subsequently, the unsigncryption and decompression processes are carried out to restore the original medical image with enhanced security. An experimental evaluation is conducted using medical chest X-ray images, to measure its performance based on aspects like peak signal-to-noise ratio (PSNR), compression ratio, space complexity, confidentiality rate, and integrity rate. The obtained results demonstrate that MWTCSELS is more efficient in achieving higher peak signal-to-noise ratios and compression ratios while maintaining strong confidentiality and utilizing less storage space compared to existing approaches.