Electrocardiography (ECG) monitor is a medical device for recording the electrical activities of the heart using electrodes placed on the body. There are many ECG monitors in the market but it is essential to find the accuracy with which they generate results. Accuracy depends on the processing of the ECG signal which contains several noises and the algorithms used for detecting peaks. Based on these peaks the abnormality in the functioning of the heart can be estimated. Hence this thesis characterizes the ECG signal which helps to detect the abnormalities and determine the accuracy of the system.

No description available.

The surface electrocardiogram (ECG) is a periodic signal portraying the electrical activity of the heart from the torso. The past fifty years have witnessed a proliferation of computer algorithms destined for ECG analysis. Signal averaging is a noise reduction technique believed to enable the surface ECG to act as a non-invasive surrogate for cardiac electrophysiology. The P wave and the QRS complex of the ECG respectively depict atrial and ventricular depolarization. QRS detection is a pre-requisite to P wave and QRS averaging. A novel algorithm for robust QRS detection in mice achieves a four-fold reduction in false detections compared to leading commercial software, while its human version boasts an error rate of just 0.29% on a public database containing ECGs with varying morphologies and degrees of noise. A fully automated P wave and QRS averaging and onset/offset detection algorithm is also proposed. This approach is shown to predict atrial fibrillation, a common cardiac arrhythmia which could cause stroke or heart failure, from normal asymptomatic ECGs, with 93% sensitivity and 100% specificity. Automated signal averaging also proves to be slightly more reproducible in consecutive recordings than manual signal averaging performed by expert users. Several studies postulated that high-frequency energy content in the signal-averaged QRS may be a marker of sudden cardiac death. Traditional frequency spectrum analysis techniques have failed to consistently validate this hypothesis. Layered Symbolic Decomposition (LSD), a novel algorithmic time-scale analysis approach requiring no basis function assumptions, is presented. LSD proves more reproducible than state-of-the-art algorithms, and capable of predicting sudden cardiac death in the general population from the surface ECG with 97% sensitivity and 96% specificity. A link between atrial refractory period and high-frequency energy content of the signal-averaged P wave is also considered, but neither LSD nor other algorithms find a meaningful correlation. LSD is not ECG-specific and may be effective in countless other signals with no known single basis function, such as other bio-potentials, geophysical signals, and socio-economic trends. / Thesis (Ph.D, Computing) -- Queen's University, 2013-09-30 23:54:21.137

There is a current trend towards the use of wearable biomedical devices for the purpose of recording various biosignals, such as electrocardiograms (ECG). Wearable devices have different issues and challenges compared to nonwearable ones, including motion artifacts and contact characteristics related to body-conforming materials. Due to this susceptibility to noise and artifacts, signals acquired from wearable devices may lead to incorrect interpretations, including false alarms and misdiagnoses. This research addresses two challenges of wearable devices. First, it investigates the effect of applied pressure on biopotential electrodes that are in contact with the skin. The pressure affects skin–electrode impedance, which impacts the quality of the acquired signal. We propose a setup for measuring skin–electrode impedance during a sequence of applied calibrated pressures. The Cole–Cole impedance model is utilized to model the skin–electrode interface. Model parameters are extracted and compared in each state of measurement with respect to the amount of pressure applied. The results indicate that there is a large change in the magnitude of skin–electrode impedance when the pressure is applied for the first time, and slight changes in impedance are observed with successive application and release of pressure. Second, this research assesses the quality of ECG signals to reduce issues related to poor-quality signals, such as false alarms. We design an algorithm based on Deep Belief Networks (DBN) to distinguish clean from contaminated ECGs and validate it by applying real clean ECG signals taken from the MIT-BIH arrhythmia database of Physionet and contaminated signals with motion artifacts at varying signal-to-noise ratios (SNR). The results demonstrate that the algorithm can recognize clean from contaminated signals with an accuracy of 99.5% for signals with an SNR of -10 dB. Once low- and high-quality signals are separated, low-quality signals can undergo additional pre-processing to mitigate the contaminants, or they can simply be discarded. This approach is applied to reduce the false alarms caused by poor-quality ECG signals in atrial fibrillation (AFib) detection algorithms. We propose a signal quality gating system based on DBN and validate it with AFib signals taken from the MIT-BIH Atrial Fibrillation database of Physionet. Without gating, the AFib detection accuracy was 87% for clean ECGs, but it markedly decreased as the SNR decreased, with an accuracy of 58.7% at an SNR of -20 dB. With signal quality gating, the accuracy remained high for clean ECGs (87%) and increased for low SNR signals (81% for an SNR of -20 dB). Furthermore, since the desired level of quality is application dependent, we design a DBN-based algorithm to quantify the quality of ECG signals. Real ECG signals with various types of arrhythmias, contaminated with motion artifacts at several SNR levels, are thereby classified based on their SNRs. The results show that our algorithm can perform a multi-class classification with an accuracy of 99.4% for signals with an SNR of -20 dB and an accuracy of 91.2% for signals with an SNR of 10 dB.

An electrocardiogram (ECG) is a bioelectrical signal which records the heart’s electrical activity versus time on the body surface via contact electrodes. The recorded ECG signal is often contaminated by noise and artifacts that can be within the frequency band of interest. This noise can hide some important features of the ECG signal. The focus of this thesis is the application of new modified versions of the Universal threshold to allow additional enhancements in the reduction of ECG noise. Despite the fact that there are many types of contaminating noises in ECG signals, only white noise and baseline wandering will be considered. This type of noise is undesirable and needs to be removed prior to any additional signal processing for proper analysis and display of the ECG signal.

An Electrocardiogram (ECG) records electrical signals from the heart to detect abnormal heart rhythms or cardiac arrhythmias. Atrial Fibrillation (AF) is the most common arrhythmia which leads to a large number of deaths annually. The diagnosis of heart disease is skill-dependent and time-consuming, therefore using an intelligent system is a time- and cost-effective approach which can also enhance diagnostic accuracy. This study uses several types of Neural Networks (NNs) including the Deep Neural Network (DNN) GoogLeNet, Multi-Layer Perceptron (MLP), Adaptive Neuro-Fuzzy Inference System (ANFIS), and Long Short-Term Memory (LSTM) to identify arrhythmias in AF signals. The results obtained are compared in order to identify the most effective and accurate system for AF diagnosis. The proposed system has two main steps, preprocessing and postprocessing. In the preprocessing step, different approaches based on the classifier network are used. More specifically, for MLP, ANFIS, and LSTM the 1-D Daubechies wavelet is used, and the extracted wavelet coefficients and statistical features are used as input data to the network. For GoogLeNet, the Continuous Wavelet Transform (CWT) is used to create a time-frequency representation of the signal (scalogram) and extract key signal features. In the postprocessing step, the data obtained (extracted features) are used as the input data to classify the signals. Also, the train and test accuracies and the running times are compared. The results obtained indicate that GoogLeNet provides the best accuracy, but its running time is long. Further, although the ANFIS and MLP networks are much faster than LSTM and GoogLeNet, their accuracy is much lower. / Graduate

The main goal of this thesis is to test the feasibility of human identification using the Electrocardiogram (ECG). Such biomedical signal has several key advantages including its intrinsic nature and liveness indicator which makes it more secure compared to some of the existing conventional and traditional biometric modalities. In compliance with the terms and regulations of McMaster University, this work has been assembled into a sandwich thesis format which consist of three journal papers. The main idea of this work is to identify individuals using distance measurement techniques and ECG feature matching. In addition, we gradually developed the content of the three papers. In the first paper, we started with the general criteria for developing ECG based biometric systems. To explain, we proposed both fiducial and non-fiducial approaches to extract the ECG features followed by providing comparative study on the performance of both approaches. Next, we applied non-overlapped data windows to extract the ECG morphological and spectral features. The former set of features include the amplitude and slope differences between the Q, R and S peaks. The later features include extracting magnitudes of the ECG frequency components using short time Fourier Transform (STFT). In addition, we proposed a methodology for QRS detection and segmentation using STFT and binary classification of ECG fiducial features. In the second paper, we proposed a technique for choosing overlapped data windows to extract the abovementioned features. Namely, the dynamic change in the ECG features from heart beats to heartbeat is utilized for identification purposes. To improve the performance of the proposed techniques we developed Frechet-mean based classifier for this application. These classifiers exploit correlation matrix structure that is not accounted for in classical Euclidean techniques. In addition to considering the center of the cluster, the Frechet-mean based techniques account for the shape of the cluster as well. In the third paper, the thesis is extended to address the variability of ECG features over multiple records. Specifically, we developed a multi-level wavelet-based filtering system which utilizes features for multiple ECGs for human identification purposes. In addition, we proposed a soft decision-making technique to combine information collected from multi-level identification channels to reach a common final class. Lastly, we evaluated the robustness of all our proposed methods over several random experiments by changing the testing data and we achieved excellent results. The results of this thesis show that the ECG is a promising biometric modality. We evaluated the performance of the proposed methods on the public ECG ID database because it was originally recorded for biometric purposes. In addition, to make performance evaluation more realistic we used two recordings of the same person obtained under possibly different conditions. Furthermore, we randomly changed both the training and testing data which are obtained from the full ECG records for performance evaluation purposes. However, it is worth mentioning that in all parts of the thesis, various parameters settings are presented to support the main ideas and it is subject to change according to human activity and application requirements. Finally, the thesis concludes with a comparison between all the proposed methods, and it provides suggestions on few open problems that can be considered for future research as extension to the work that has been done in this thesis. Generally, these problems are associated with the constraints on computational time, data volume and ECG clustering. / Thesis / Doctor of Philosophy (PhD)

Physicians’ understanding of bio-signals, measured using medical instruments, becomes the foundation of their decisions and diagnoses of patients, as they rely strongly on what the instruments show. Thus, it is critical and very important to ensure that the instruments’ readings exactly reflect what is happening in the patient’s body so that the detected signal is the real one or at least as close to the real in-body signal as possible and carries all of the appropriate information. This is such an important issue that sometimes physicians use invasive measurements in order to obtain the real bio-signal. Generating an in-body signal from what a measurement device shows is called “signal purification” or “reconstruction,” and can be done only when we have adequate information about the interface between the body and the monitoring device. In this research, first, we present a device that we developed for electrocardiogram (ECG) acquisition and transfer to PC. In order to evaluate the performance of the device, we use it to measure ECG and apply conductive textile as our ECG electrode. Then, we evaluate ECG signals captured by different electrodes, specifically traditional gel Ag/AgCl and dry golden plate electrodes, and compare the results. Next, we propose a method to reconstruct the ECG signal from the signal we detected with our device with respect to the interface characteristics and their relation to the detected ECG. The interface in this study is the skin-electrode interface for conductive textiles. In the last stage of this work, we explore the effects of pressure on skin-electrode interface impedance and its parametrical variation.

Physicians’ understanding of bio-signals, measured using medical instruments, becomes the foundation of their decisions and diagnoses of patients, as they rely strongly on what the instruments show. Thus, it is critical and very important to ensure that the instruments’ readings exactly reflect what is happening in the patient’s body so that the detected signal is the real one or at least as close to the real in-body signal as possible and carries all of the appropriate information. This is such an important issue that sometimes physicians use invasive measurements in order to obtain the real bio-signal. Generating an in-body signal from what a measurement device shows is called “signal purification” or “reconstruction,” and can be done only when we have adequate information about the interface between the body and the monitoring device. In this research, first, we present a device that we developed for electrocardiogram (ECG) acquisition and transfer to PC. In order to evaluate the performance of the device, we use it to measure ECG and apply conductive textile as our ECG electrode. Then, we evaluate ECG signals captured by different electrodes, specifically traditional gel Ag/AgCl and dry golden plate electrodes, and compare the results. Next, we propose a method to reconstruct the ECG signal from the signal we detected with our device with respect to the interface characteristics and their relation to the detected ECG. The interface in this study is the skin-electrode interface for conductive textiles. In the last stage of this work, we explore the effects of pressure on skin-electrode interface impedance and its parametrical variation.

The fetal electrocardiogram contains within it, information regarding the health of the fetus. Currently, fetal ECG is recorded directly from the scalp of the baby during labour. However, it has been shown that fetal ECG can also be measured using surface electrodes attached to a pregnant mother's abdomen. The advantage of this method lies in the fact that fetal ECG can be measured noninvasively before the onset of labour. The difficulty lies in isolating the fetal ECG from extraneous signals that are simultaneously recorded with it. Several signal processing methodologies have been put forth in order to extract the fetal ECG component from a mixture of signals. Two recent techniques that have been put forth include a scheme that has previously been used to nonlinearly reduce noise in deterministically chaotic noise and the other uses a blind source separation technique called independent component analysis. In this thesis, we describe the significance of the fetal electrocardiogram as a diagnostic tool in medicine, a brief overview of the theory behind the nonlinear noise reduction technique and blind source separation, and results from having processed synthetic and real data using both techniques. We find that although the noise reduction technique performs adequately, the blind source separation process performs faster and more robustly against similar data. The two techniques can be used in tandem to arrive at an approximate fetal ECG signal, which can be further analyzed by calculating, for example, the fetal heart rate. / Thesis / Master of Engineering (ME)