Enhancing Privacy of information with Data Embedding in Medical MRI Images Based on Segmentation and HVS Model

Abstract

The development of communications technology and the arrival of Medicine modern equipment in the health domain has caused the diagnosis and treatment methods to be considered from a distance and medical centers are equipped with telemedicine systems. Forensic medicine organizations with daily clients and outpatient examinations such as accident clients, Conflict returns as well as spousal abuse. Doctors and employees of the organization also serve as one of the powerful arms of the judiciary, following up on important cases in the medical, laboratory, and psychiatric commissions So that they can take steps to realize the rights of the people. Patient data security has become a serious concern for professionals and one of the methods is using data embedding to the protection against these risks. In this method, medical informatics, telemedicine, and forensic medicine organizations has played a pivotal role and any mistake in the reporting can be catastrophic. The main purpose of this research is to present data on EPRs with Enhancing data embedding Based on SSIM and HVS with the help of medical image segmentation and focus on brain MRI images. In this study, innovations include the addition of the HVS block based on the SSIM criterion to meet transparency and robustness conditions. Selection of the embedding coefficient K is considered adaptively depending on the degree of uniformity of the N - ROI region with the image quality factor. The coordinates of ROI areas in one of the DCT and DWT conversion blocks have been demonstrated to have better performance at Concealment EPRs. The choice of coefficients af,k which consists of the optimization frequency sensitivity function and spatial property is comparatively done to match the visual perception of the visual system. The present study aims to improve the effectiveness of the proposed method, improve the security level, and the confidentiality of patient information, and integrate the storage of patient information and image. The simulation results of the proposed method, considering the parameters of embedding and transparency in comparison with other methods, have been done using evaluation criteria including MSE PSNR, NC, SSIM, and BER.