N-Grams Based A Novel Adapted Watermarking Approach to Secure Health Information Systems Databases: Case of Medical Images

Abstract

In regards to data protection, securing medical content must achieve a balance between several factors such as data confidentiality, data integrity, data availability, etc. Health Information Systems (HIS) has been identified for a few years, since establishments have largely computerized their healthcare process and biomedical devices have been developed incorporating complex computer systems. Access to patient information has become purely electronic through Electronic Patient Record (EPR). These records contain information in different formats (textual, images, videos). In the case of securing medical images which is the subject of this paper, it is imperative to find solutions with a tradeoff between the security degree and its complexity in terms of computation time. In this paper, we propose a new approach for medical images watermarking that addresses the problem described above by preserving an acceptable tradeoff between security and complexity. This approach is based on the N-grams technique, with the aim to watermark only the regions of interest (ROI) within the host image. This approach has proven its efficiency in Natural Language Processing (NLP). The results obtained are very encouraging and will be detailed in this paper in terms of imperceptibility, execution time and robustness. For the imperceptibility most of PSNR results are well over 40dB. Regarding the robustness, the values of the normalized correlation coefficients are mostly close to 1. According to these two metrics we can conclude that the proposed watermarking scheme is well imperceptible and robust.

Keywords:

Medical images N-grams Region of interest imperceptibility robustness

Downloads

Download data is not yet available.
LAOUAMER, L. (2024). N-Grams Based A Novel Adapted Watermarking Approach to Secure Health Information Systems Databases: Case of Medical Images. JOURNAL OF ADMINISTRATIVE AND ECONOMIC SCIENCES, 17(1), 87–100. Retrieved from https://jaes.qu.edu.sa/index.php/jae/article/view/2451
Copyright and license info is not available
Copyright and license info is not available
Author biographies is not available.