In this paper, we present a novel blind watermarking 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.
This paper presents the results of morphological heart arrhythmia detection based on features of electrocardiography, ECG, signal. These signals are obtained from MIT/BIH arrhythmia database. The ECG beats were first modeled using Hermitian basis functions, (HBF). In this step, the width parameter, sigma, of HBF was optimized to minimize the model error. Then, the feature vector which consists of the parameters of the model is used as an input to k-nearest neighbor, kNN, classifier to examine the efficiency of the model. In our experiments, seven different types of arrhythmias have been considered. We achieved the sensitivity of 99.00% and specificity of 99.84% which are comparable to previous works. These results were obtained in less than 0.6 second which is suitable for real-time diagnosis of heart arrhythmias.
BackgroundElectrocardiography (ECG) signal is a primary criterion for medical practitioners to diagnose heart diseases. The development of a reliable, accurate, non-invasive and robust method for arrhythmia detection could assists cardiologists in the study of patients with heart diseases. This paper provides a method for morphological heart arrhythmia detection which might have different shapes in one category and also different morphologies in relation to the patients. The distinctive property of this method in addition to accuracy is the robustness of that, in presence of Gaussian noise, time and amplitude shift.MethodsIn this work 2nd, 3rd and 4th order cumulants of the ECG beat are calculated and modeled by linear combinations of Hermitian basis functions. Then, the parameters of each cumulant model are used as feature vectors to classify five different ECG beats namely as Normal, PVC, APC, RBBB and LBBB using 1-Nearest Neighborhood (1-NN) classifier. Finally, after classifying each model, a final decision making rule apply to these specified classes and the type of ECG beat is defined.ResultsThe experiment was applied for a set of ECG beats consist of 9367 samples in 5 different categories from MIT/BIH heart arrhythmia database. The specificity of 99.67% and the sensitivity of 98.66% in arrhythmia detection are achieved which indicates the power of the algorithm. Also, the accuracy of the system remained almost intact in the presence of Gaussian noise, time shift and amplitude shift of ECG signals.ConclusionsThis paper presents a novel and robust methodology in morphological heart arrhythmia detection. The methodology based on the Hermite model of the Higher-Order Statistics (HOS). The ability of HOS in suppressing morphological variations of different class-specific arrhythmias and also reducing the effects of Gaussian noise, made HOS, suitable for detection morphological heart arrhythmias. The proposed method exploits these properties in conjunction with Hermitian model to perform an efficient and reliable classification approach to detect five morphological heart arrhythmias. And the time consumption of this method for each beat is less than the period of a normal beat.
In this paper, a new piecewise modeling for approximation of ECG signal is presented. Most of the modeling methods are focused to obtain the best approximation of the entire ECG signal. The proposed method exploits the importance of different intervals of ECG signals, in particular QRS complex, by performing a segmented based modeling using Hermitian basis functions. This yields to weighting the approximation error of each segment based on its importance throughout the ECG complex. As the result shows the total error obtained in this method is almost halved in comparison with similar non-segmented method. This has a great impact in modeling the heart arrhythmias where a small error could mislead the diagnosis. The presented method uses only the 5th order Hermitian basis functions which considerably reduce the total parameters needed to represent the ECG signal in comparison with other Hermitian based methods.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.