Wrapping based curvelet transform approach for ECG watermarking in telemedicine application. (May 2022)

• Record Type: Journal Article
• Authors: Khaldi, Amine; Kafi, Med Redouane; Moad, Med Sayah
• Is Part Of: Biomedical signal processing and control. Volume 75(2022)
• Publication Date: 2022-05
• Subjects: Telemedicine -- Watermarking -- Electrocardiogram signals -- Quick response code -- Schur decomposition -- Fast discrete curvelet transform; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2022.103540
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: A watermark composed of the patient's information and ECG acquisition parameters are concealed in ECG signal. To extract the image's frequency content, a fast discrete curvelet transform is performed which allows obtaining a better multi-scale representation than those provided by wavelets. In the proposed blind approach, the extraction of the mark is done without using the original signal. The tests demonstrated that our approach allows insertion of a large amount of data while preserving a reasonable imperceptibility. Abstract: Patient information is currently maintained separately from the physiological signals received in telemedicine. Furthermore, in recent years, the usage of portable devices to obtain such medical information has become more common. This form of storing can lead to several other authentication issues and compromise the accuracy of diagnostics. To address this issue, we propose a watermarking method for embedding patient information in the ECG signal in this study. In our technique, the watermark is a QR code that contains the patient's information as well as the Electrocardiogram signal. The 1D ECG signal is transformed to a 2D signal in the proposed technique. A fast discrete curvelet transform is used to extract the signal's frequency content, and then Schur decomposition is used to the frequency sub-bands obtained. Finally, the watermark bits are integrated by modifying the Eigen values produced. According to imperceptibility experiments, this method produces a watermarked signal that is identical to the originals, keeping the diagnostic content of the signal. Robustness test indicated that the watermark could overcome the most popular watermarking attacks.