A novel medical image watermarking technique for detecting tampers inside ROI and recovering original ROI

Abstract

Transmission of medical images among remote places is a general practice in telemedicine. Medical images may be modified intentionally or accidentally as the transmission of these may take place through unsecure networks like internet. Before making any diagnostic decisions, the medical practitioner has to verify the integrity of Region of Interest (ROI) in the received medical image in order to avoid wrong diagnosis. Watermarking can be used for checking the integrity of medical images. In this paper, we propose a novel medical image watermarking method based on Integer Wavelet Transform (IWT). This proposal verifies the integrity of ROI, precisely identifies tampered blocks inside ROI, provide robustness to the data embedded inside RONI and recover original ROI without loss. In the proposed method, the medical image is segmented into three regions: ROI, RONI (Region of Non Interest) and border. Hash value of ROI, information for identifying the ROI region and patient's data are embedded inside border pixels. Recovery data of ROI is embedded into RONI using IWT. Experimental results shows that the proposed method provides robustness to the ROI recovery data embedded inside RONI and accurately detects and localizes tampered areas inside ROI and recovers the original ROI exactly.