A Novel Blind Watermarking of ECG Signals on Medical Images Using EZW Algorithm Mohammad S. Nambakhsh, Alireza Ahmadian*, Mohammad Ghavami, Reza S.Dilmaghani and S. Karimi-Fard Abstract—In this paper, we present a novel blind water- marking method with secret key by embedding ECG signals in medical images. The embedding is done when the original image is compressed using the embedded zero-tree wavelet (EZW) algorithm. The extraction process is performed at the decompression time of the watermarked image. Our algorithm has been tested on several CT and MRI images and the peak signal to noise ratio (PSNR) between the original and watermarked image is greater than 35 dB for watermarking of 512 to 8192 bytes of the mark signal. The proposed method is able to utilize about 15% of the host image to embed the mark signal. This marking percentage has improved previous works while preserving the image details. I. INTRODUCTION Exchange of database between hospitals needs efficient and reliable transmission and storage techniques to cut down the cost of health care. This exchange involves large amount of vital patient information such as bio-signals and medical images. Interleaving one form of data such as 1- D signal, over digital images can combine the advantages of data security with efficient memory utilization [1], but nothing prevents the user from manipulating or copying the decrypted data for illegal uses. Embedding vital information of patients inside their scan images will help physics to make a better diagnosis of disease. In order to solve these issues, watermark algorithms have been proposed as a way to complement the encryption processes and provide some tools to track the retransmission and manipulation of multimedia contents [2][3]. A watermarking system is based on an imperceptible insertion of a watermark (a signal) in an image. This technique is adapted here for interleaving graphical ECG signals within medical images, to reduce storage and M. S. Nambakhsh is with the Department of Biomedical Systems & Medical Physics, Tehran University of Medical Sciences, and Research Center for Science and Technology in Medicine, TUMS, Tehran, Iran nambakhsh@razi.tums.ac.ir A. Ahmadian (corresponding author) is an Associate Professor at the Tehran University of Medical Sciences, TUMS, Dept. of Biomedical Sys- tems & Medical Physics and Research Center for Science and Technology in Medicine, TUMS, Tehran, Iran ahmadian@sina.tums.ac.ir M. Ghavami is a Reader in UWB Communications with the Ultra Wideband Communication Research, King’s College London, University of London, UK mohammad.ghavami@kcl.ac.uk R. S. Dilmaghani is a Lecturer with the Ultra Wideband Commu- nication Research, King’s College London, University of London, UK reza.shams dilmaghani@kcl.ac.uk S. Karimi-Fard is with the Department of Biomedical Systems & Medical Physics, Tehran University of Medical Sciences, and Research Center for Science and Technology in Medicine, TUMS, Tehran, Iran karimifard@razi.tums.ac.ir transmission overheads as well as helping for computer aided diagnostics system. In this paper we present a new watermarking method combined with the EZW-based wavelet coder. The principle is to replace significant wavelet coefficients of ECG signals by the corresponding significant wavelet coefficients belong- ing to the host image which is much bigger in size than the mark signal. This paper presents a brief introduction to watermarking and the EZW coder that acts as a platform for our watermarking algorithm. II. EZW IN IMAGE The EZW algorithm was originally developed by Shapiro [4] to find the best transmission order of the wavelet coeffi- cients which is the absolute value of decreasing order. This algorithm has already been applied to medical images and the electrocardiogram with good success [1]. The wavelet transform is a dyadic decomposition of an im- age [5] achieved by a pair of quadratic mirror filters (QMF). In two-dimensional separable dyadic discrete wavelet trans- form (DWT), each level of decomposition produces four bands of data, one corresponding to the low pass band (LL), and the other three corresponding to horizontal (HL), vertical (LH), and diagonal (HH) high pass bands. Fig. 1 shows this concept. Fig. 1. levels of decomposition. Two distinct properties of the EZW algorithm make it an effective means of compression, as compared with traditional approaches. Firstly, the EZW algorithm exploits the hierar- chy of the wavelet coefficients, and establishes a connection between coefficients from different sub-bands, allowing mul- tiple coefficients to be encoded simultaneously. Secondly, coefficients are encoded in order of importance using bit prioritization [1]. After performing the 2-D DWT on the image, the resulting wavelet coefficients are coded by using a decreasing sequence of thresholds, that is T 0 ,T 1 ,...,T N-1 : T i = T i-1 2 and T 0 =2 log 2 (max{|γ(x,y)|}) (1) Proceedings of the 28th IEEE EMBS Annual International Conference New York City, USA, Aug 30-Sept 3, 2006 FrC10.2 1-4244-0033-3/06/$20.00 ©2006 IEEE. 3274