Curvelets-based ECG steganography for data security

Biomedical signals transmitted over the internet are usually tagged with patient information. Data hiding techniques such as steganography ensures the security of such data by hiding the data into signals. However, data hiding results in signal deterioration that might affect diagnosability. A novel technique which uses curvelet transforms to hide patient information into their ECG signal is presented. Curvelet transform decomposes the ECG signal into frequency sub-bands. A quantisation approach is used to embed patient data into coefficients whose values are around zero, in the high-frequency sub-band. Performance metrics provide the measure of watermark imperceptibility of the proposed approach. BER is used to measure the ability to extract patient data. The proposed approach is demonstrated on the MIT-BIH database and the observations validate that its performance is superior compared with the random locations approach. Although the performance of the proposed approach decreases as patient information size increases, the peak signal-to-noise ratio values are high. Therefore, the proposed approach can be used for the safe transfer of patient data.