Affiliations 

  • 1 Faculty of Computer Systems and Software Engineering, Universiti Malaysia Pahang, Lebuhraya Tun Razak, 26300 Gambang, Kuantan, Pahang Darul Makmur, Malaysia. [email protected]
  • 2 Faculty of Computer Systems and Software Engineering, Universiti Malaysia Pahang, Lebuhraya Tun Razak, 26300 Gambang, Kuantan, Pahang Darul Makmur, Malaysia
J Digit Imaging, 2017 Jun;30(3):328-349.
PMID: 28050716 DOI: 10.1007/s10278-016-9930-9

Abstract

Tampering on medical image will lead to wrong diagnosis and treatment, which is life-threatening; therefore, digital watermarking on medical image was introduced to protect medical image from tampering. Medical images are divided into region of interest (ROI) and region of non-interest (RONI). ROI is an area that has a significant impact on diagnosis, whereas RONI has less or no significance in diagnosis. This paper has proposed ROI-based tamper detection and recovery watermarking scheme (ROI-DR) that embeds ROI bit information into RONI least significant bits, which will be extracted later for authentication and recovery process. The experiment result has shown that the ROI-DR has achieved a good result in imperceptibility with peak signal-to-noise ratio (PSNR) values approximately 48 dB, it is robust against various kinds of tampering, and the tampered ROI was able to recover to its original form. Lastly, a comparative table with the previous research (TALLOR and TALLOR-RS watermarking schemes) has been derived, where these three watermarking schemes were tested under the same testing conditions and environment. The experiment result has shown that ROI-DR has achieved speed-up factors of 22.55 and 26.65 in relative to TALLOR and TALLOR-RS watermarking schemes, respectively.

* Title and MeSH Headings from MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine.