Evaluation of an automatic threshold based detector of waveform limits in Holter ECG with the QT database

Jané, R.; Blasi, Anna; García, J.; Laguna, Pablo

doi:10.1109/cic.1997.647889

Cited by 78 publications

(42 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Jane [22] has proposed a general framework to analyse the results of delineator algorithms evaluated on a given database, such as the QTDB. Our results are issued from the same framework and compared to those achieved by [2] and [21] (a low-pass-differentiator-based algorithm).…”

Section: B Delineation Resultsmentioning

confidence: 99%

Improving ECG Beats Delineation With an Evolutionary Optimization Process

Dumont

Hernández

Carrault

2010

IEEE Trans. Biomed. Eng.

View full text Add to dashboard Cite

Abstract-As in other complex signal processing tasks, ECG beat delineation algorithms are usually constituted of a set of processing modules, each one characterized by a certain number of parameters (filter cutoff frequencies, threshold levels, time windows...). It is well recognized that the adjustment of these parameters is a complex task that is traditionally performed empirically and manually, based on the experience of the designer. In this work, we propose a new automated and quantitative method to optimize the parameters of such complex signal processing algorithms. To solve this multiobjective optimization problem, an evolutionary algorithm (EA) is proposed. This method for parameter optimization is applied to a Wavelet-Transform-based ECG delineator that has previously shown to present interesting performances. An evaluation of the final delineator, using the optimal parameters, has been performed on the QT database from Physionet and results are compared with previous algorithms reported in the literature. The optimized parameters provide a more accurate delineation, with a global improvement, over all the criteria evaluated, and over the best results find in the literature, measured of 7.7%, which is a proof of the interest of the approach.

show abstract

Section: B Delineation Resultsmentioning

confidence: 99%

Improving ECG Beats Delineation With an Evolutionary Optimization Process

Dumont

Hernández

Carrault

2010

IEEE Trans. Biomed. Eng.

View full text Add to dashboard Cite

show abstract

“…It consists on the application of signal processing methods to the ECG signals in order to identify the beats in the ECG signals and the boundaries of the waves that compose them and to extract the parameter values that are relevant for the rest of the algorithm. For this task we have used the ECGPUWAVE tool [4], a threshold based ECG wave boundaries detector that we previously used successfully in real time ECG analysis applications [5]. For each beat the fiducial point of the QRS, the iso-electric point (considered as the beginning of the QRS complex) and the J-point (considered as the end of QRS complex) are identified.…”

Section: Preprocessingmentioning

confidence: 99%

“…The filter applied to the ECG signal is the same used by the ECGPUWAVE tool (a Lynn filter) [4]. The iso-electric and ST level have been averaged over the two adjacent samples in order to minimize the 50/60 Hz noise.…”

Section: Preprocessingmentioning

confidence: 99%

An on-line ischemia monitoring algorithm for mobile devices

Dranca

Goñi

Rodríguez

et al. 2005

Computers in Cardiology, 2005

View full text Add to dashboard Cite

The early diagnosis of ischemic events may prevent irreversible damage to the heart muscle. Mobile IntroductionMyocardial ischemia is one of the diseases with highest incidence rate in the industrialised countries. Physiologically, is identified by an insufficient oxygenated blood supply compared to current myocardial demand. This event is reflected in a ECG signal as anomalous variations during the ventricular re-polarization. Although the ECG signal analysis is not the most accurate method that exists to detect the ischemic events, it is without question, the least invasive and costly one, and still maintain a high sensitivity level in the detection.Moreover, prolonged, severe or repeated ischemic episodes can lead to irreversible damages to the cardiac tissue. Therefore it comes up the importance of the early detection of such kind of episodes.The latest technological advances in the communication and mobile devices have led to significant progresses of the mobile computing area and give the possibility of new any time and anywhere telemonitoring solution. The combination of real time ischemia detection methods with mobile computing techniques may give the solution to the early detection of ischemic events through innovative telemonitoring systems.The aim of this paper is to present the structure and the validation results of a detection algorithm of transient ischemic events based on ECG signal analysis. The algorithm can be embedded in a mobile such as a PDA and can be executed in real time. Those requirements involve some rules that we had in mind during the development of the algorithm: 1. to be simple enough and with a low computation cost to fulfill the time restrictions of a real time execution in a device with limited computation capabilities and memory capacity (PDA); 2. to generate a minimal number of false alarms 3. to be highly sensitive to ischemic episodes and above all to the most dangerous ones. We start this paper by mentioning some related work, followed by a description of the transient ischemia detection algorithm. We finish presenting the performances of the algorithm and conclusions. Related work and materialsThe ischemic transient events are associated with variations (elevations or depressions) of the ST segment in the ECG signal. Nevertheless, similar variations may be also produced by diurnal changes, postural changes, changes in ventricular conduction which make difficult the distinction between ischemic and non-ischemic events. There are several distinct algorithms that address the automatically ischemic detection in the ECG signals applying different methods that consider time domain analysis, KL transform, neural network or fuzzy logic. Nevertheless only a few of them deal explicitly with non-ischemic events such as axis shift events (an interesting reference of the existing ischemia algorithms can be found in [1]). Moreover the majority of the existing algorithms are difficult to be adjusted for real time execution and up to our knowledge, only a few algorithms specia...

show abstract

“…We extracted the subset of MIMIC II that 1) contained ECG waveforms matched to ICD-9 codes and 2) corresponded to one of the four disease etiologies modelled below -congestive heart failure (CHF), primary cardiomyopathy (PCM), and chronic pulmonary heart disease (CPHD) and CAD. From each subject's corresponding waveforms, up to 10 R-R time series, each with one minute duration, were extracted using the most recent version of PhysioNet's ecgpuwave beat detector [17]. In total, there were 613 congestive heart failure, 74 cardiomyopathy, and 71 chronic pulmonary disease patients in MIMIC II, and 271 patients in the CAD data set.…”

Section: Methodsmentioning

confidence: 99%

Network Analysis of Heart Beat Intervals Using Horizontal Visibility Graphs

Madl

2016

2016 Computing in Cardiology Conference (CinC)

View full text Add to dashboard Cite

Heart beat interval time series contain information predictive of heart disease, but most current predictors do not provide sufficient reliability for clinical use. Using several predictors improves predictive power, but the limit is not yet known, suggesting that not all the information in interbeat interval series has been captured by previous work.We convert heart beat time series into scale-free networks using horizontal visibility graphs (HVGs), which are well-suited to distinguishing deterministic dynamical systems from stochastic systems, allowing them to model new aspects of autonomic heart rate modulation. Based on the HVG, we introduce and evaluate a general class of predictors, which can be used to augment existing features used in heart rate variability (HRV) analysis, and which exhibit high predictive power for several types of heart disease.We show the statistical significance of these network predictors, and their competitive performance to popular statistical, geometric and non-linear features, on ICU and Holter ECGs, including several heart disease etiologies.

show abstract

Evaluation of an automatic threshold based detector of waveform limits in Holter ECG with the QT database

Cited by 78 publications

References 4 publications

Improving ECG Beats Delineation With an Evolutionary Optimization Process

Improving ECG Beats Delineation With an Evolutionary Optimization Process

An on-line ischemia monitoring algorithm for mobile devices

Network Analysis of Heart Beat Intervals Using Horizontal Visibility Graphs

Contact Info

Product

Resources

About