Protecting patient confidential information based on ECG reversible data hiding

Abstract

Nowadays telecardiology is widely popular due to the fact that an increasing number of people are suffering from cardiac disease in the world. Therefore huge amount of ECG signal as well as patient confidential information will be transmitted via the Internet. Ibaida’s wavelet-based data hiding technique aims to protect patient confidential data utilizing ECG signal as a host media. But it cannot completely reconstruct the original ECG signal. Any alteration of the ECG may lead to an inaccurate diagnosis conclusion drawn by the doctor, which cannot be accepted by patients. In this paper, our elemental standpoint requires that both patient information and ECG signal must be perfectly restored at the extraction side. Firstly a method is proposed to embed patient confidential data into ECG signal, while keeping its high visual quality. Then we use a unified embedding-scrambling method to guarantee the security of patient privacy as well as the ECG signal itself. The structure of watermarked ECG signal is severely deconstructed. Both of the experimental results demonstrate that our proposed methods are reversible. Moreover the latter scheme can achieve high information payload.