In this paper, we developed and evaluated a robust single-lead electrocardiogram (ECG) delineation system based on the wavelet transform (WT). In a first step, QRS complexes are detected. Then, each QRS is delineated by detecting and identifying the peaks of the individual waves, as well as the complex onset and end. Finally, the determination of P and T wave peaks, onsets and ends is performed. We evaluated the algorithm on several manually annotated databases, such as MIT-BIH Arrhythmia, QT, European ST-T and CSE databases, developed for validation purposes. The QRS detector obtained a sensitivity of Se=99.66% and a positive predictivity of P+=99.56% over the first lead of the validation databases (more than 980,000 beats), while for the well-known MIT-BIH Arrhythmia Database, Se and P+ over 99.8% were attained. As for the delineation of the ECG waves, the mean and standard deviation of the differences between the automatic and manual annotations were computed. The mean error obtained with the WT approach was found not to exceed one sampling interval, while the standard deviations were around the accepted tolerances between expert physicians, outperforming the results of other well known algorithms, especially in determining the end of T wave.

A wavelet-based ECG delineator: evaluation on standard databases

https://www.repositorio.ufop.br/bitstream/123456789/7011/1/ARTIGO_ECGHeartBeat.pdf

ECG-based heartbeat classification for arrhythmia detection

QRS and ventricular beat detection is a basic procedure for electrocardiogram (ECG) processing and analysis. Large variety of methods have been proposed and used, featuring high percentages of correct detection. Nevertheless, the problem remains open especially with respect to higher detection accuracy in noisy ECGs A real-time detection method is proposed, based on comparison between absolute values of summed differentiated electrocardiograms of one of more ECG leads and adaptive threshold. The threshold combines three parameters: an adaptive slew-rate value, a second value which rises when high-frequency noise occurs, and a third one intended to avoid missing of low amplitude beats. Two algorithms were developed: Algorithm 1 detects at the current beat and Algorithm 2 has an RR interval analysis component in addition. The algorithms are self-adjusting to the thresholds and weighting constants, regardless of resolution and sampling frequency used. They operate with any number L of ECG leads, self-synchronize to QRS or beat slopes and adapt to beat-to-beat intervals. The algorithms were tested by an independent expert, thus excluding possible author's influence, using all 48 full-length ECG records of the MIT-BIH arrhythmia database. The results were: sensitivity Se = 99.69 % and specificity Sp = 99.65 % for Algorithm 1 and Se = 99.74 % and Sp = 99.65 % for Algorithm 2. The statistical indices are higher than, or comparable to those, cited in the scientific literature.

/pdf/real-time-electrocardiogram-qrs-detection-using-combined-2axkdtaf2n.pdf

Real time electrocardiogram QRS detection using combined adaptive threshold

The emergence of wireless technologies and advancements in on-body sensor design can enable change in the conventional health-care system, replacing it with wearable health-care systems, centred on the individual. Wearable monitoring systems can provide continuous physiological data, as well as better information regarding the general health of individuals. Thus, such vital-sign monitoring systems will reduce health-care costs by disease prevention and enhance the quality of life with disease management. In this paper, recent progress in non-invasive monitoring technologies for chronic disease management is reviewed. In particular, devices and techniques for monitoring blood pressure, blood glucose levels, cardiac activity and respiratory activity are discussed; in addition, on-body propagation issues for multiple sensors are presented.

Detecting Vital Signs with Wearable Wireless Sensors

Correct detection and classification of ventricular fibrillation (VF) and rapid ventricular tachycardia (VT) is of pivotal importance for an automatic external defibrillator and patient monitoring. In this paper, a VF/VT classification algorithm using a machine learning method, a support vector machine, is proposed. A total of 14 metrics were extracted from a specific window length of the electrocardiogram (ECG). A genetic algorithm was then used to select the optimal variable combinations. Three annotated public domain ECG databases (the American Heart Association Database, the Creighton University Ventricular Tachyarrhythmia Database, and the MIT-BIH Malignant Ventricular Arrhythmia Database) were used as training, test, and validation datasets. Different window sizes, varying from 1 to 10 s were tested. An accuracy (Ac) of 98.1%, sensitivity (Se) of 98.4%, and specificity (Sp) of 98.0% were obtained on the in-sample training data with 5 s-window size and two selected metrics. On the out-of-sample validation data, an Ac of 96.3% ± 3.4%, Se of 96.2% ± 2.7%, and Sp of 96.2% ± 4.6% were obtained by fivefold cross validation. The results surpass those of current reported methods.

Ventricular Fibrillation and Tachycardia Classification Using a Machine Learning Approach

A QRS complex detection algorithm was developed using the available leads of the electrocardiogram (ECG). This detector is based on the combination of two improved versions of QRS detectors available in the literature. An important characteristic of this algorithm is the possibility of using two or more ECG channels for QRS detection. The first detection method is based on a cross number in a detection threshold defined by the authors. When a low reliability situation occurs in the first method, the output of the second detection method is used to confirm or reject the detection. The second method also uses an adaptive detection threshold defined by the authors and a candidate QRS is tested against some criteria that use features as amplitude, width and RR interval to validate the candidate as a QRS. Testing the algorithm with MIT/BIH Arrhythmia Database resulted in 99.22% sensitivity and 99.73% positive predictivity.

A QRS complex detection algorithm using electrocardiogram leads

The main difficulty to determine the systolic, mean and diastolic blood pressure using the oscillometric method is related to the strategy involved in the computational algorithm to extract these parameters from the oscillometric pulse profiles. A new strategy determined correlating several quantities, such as reference blood pressure measurements, actual cuff pressure, pulse amplitude, characteristic ratios, age, weight, height, arm circumference size, systolic, mean and diastolic blood pressures, is presented in this paper. The results were obtained from an extensive statistical analysis performed with MATLAB to determine the cross-relations and distributions between characteristic ratios and the several parameters studied. The strategy was used in an oscillometric blood pressure measurement system with very good results, according to a comparative study using a simulator, providing reduction in measurement time and larger rejection of motion artifact and arrhythmia.

http://ieeexplore.ieee.org/iel5/7213/19426/00898494.pdf

A strategy for determination of systolic, mean and diastolic blood pressures from oscillometric pulse profiles

The assessment of the mechanical properties of the respiratory system is typically done by oscillating flow into the lungs via the trachea, measuring the resulting pressure generated at the trachea, and relating the two signals to each other in terms of some suitable mathematical model. If the perturbing flow signal is broadband and not too large in amplitude, linear behavior is usually assumed and the input impedance calculated. Alternatively, some researchers have used flow signals that are narrow band but large in amplitude, and invoked nonlinear lumped-parameter models to account for the relationship between flow and pressure. There has been little attempt, however, to deal with respiratory data that are both broadband and reflective of system nonlinearities. In the present study, we collected such data from mice. To interpret these data, we first developed a time-domain approximation to a widely used model of respiratory input impedance. We then extended this model to include nonlinear resistive and elastic terms. We found that the nonlinear elastic term fit the data better than the linear model or the nonlinear resistance model when amplitudes were large. This model may be useful for detecting overinflation of the lung during mechanical ventilation.

/pdf/nonlinear-and-frequency-dependent-mechanical-behavior-of-the-2xmlc9xnnw.pdf

Nonlinear and frequency-dependent mechanical behavior of the mouse respiratory system.

This paper presents a new method for detecting Ventricular Fibrillation (VF) based on a two step processing. The first one, based on the published VF Filter Leakage Method, consists in sampling data from the ECG and filtering it so that the more the signal resembles a sinusoidal wave, the lower the filter's output value will be. This idea comes from the VF wave inspection which shows that it has some sinusoidal characteristics. In order to avoid false positive occurrences due to Ventricular Tachycardia (VT) episodes (which are sinusoidal as well), a second step, based on the Complexity Measure algorithm, was used consisting in generating a binary sequence by comparison between the sampled data and an adjustable threshold, and then, using this sequence to calculate the signal's disorder level. The achieved sensitivity for the VF algorithm was 70.32% and its predictivity was 94.66%.

Ventricular fibrillation detection using a leakage/complexity measure method

The problem of interference on pacemakers as a result of extraneous electric current and electromagnetic fields is important since more than 1.5 million people worldwide have pacemakers implanted. Each year at least 100,000 people have cardiac pacemakers implanted. This article reports several effects of two electronic autodefense devices on some cardiac pacemakers. The results show that the interference affects severely the behavior of the cardiac pacemakers, but if the duration is no more than 5 s, the effects are not permanent, and the pacemakers return to the correct action as soon as the interference ceases.

José Carlos Teixeira de Barros Moraes

Papers

A QRS complex detection algorithm using electrocardiogram leads

A strategy for determination of systolic, mean and diastolic blood pressures from oscillometric pulse profiles

Nonlinear and frequency-dependent mechanical behavior of the mouse respiratory system.

Ventricular fibrillation detection using a leakage/complexity measure method

Effects of electronic autodefense devices on cardiac pacemakers.