A NOVEL APPROACH TO MEDICAL IMAGE WATERMARKING FOR TAMPER DETECTION AND RECOVERY OF REGION OF INTEREST USING PREDICTIVE CODING AND HASHING

摘要

Telemedicine is an important application based on telecommunication and information technology to provide medical services to the remote patients with the advice of the healthcare professionals. The treatment is planned by analyzing the medical reports like CT scan, MRI scan, X-ray images etc. Hence, transmission of these images over internet needs integrity and security. Medical image watermarking has gained tremendous importance due to its catering needs in diagnosis, treatment planning, and to increase the security, confidentiality and integrity of the medical images. In medical image watermarking, the significant part of the medical image is treated as Region of Interest (ROI) that is important for diagnosis. Hence, it is very important to maintain integrity and transmit the ROI securely. In this paper, predictive coding and hashing technique is applied to ROI region in order to embed in Region of Non-Interest (RONI) to generate watermarked image. The proposed method, effectively extracts the ROI at the receiver side in spite of the tampering and noise added to it. The experimental results of the proposed method on various medical image databases proved to produce robust medical images to tampering. Compared with the existing techniques, the proposed method outperforms with the Peak Signal to Noise Ratio (PSNR) approximately 50dB.