Irena Jekova

Premature ventricular contraction classification by the Kth nearest-neighbours rule

An analysis of electrocardiographic pattern recognition parameters for premature ventricular contraction (PVC) and normal (N) beat classification is presented. Twenty-six parameters were defined: 11 x 2 for the two electrocardiogram (ECG) leads, width of the complex and three parameters derived from a single-plane vectorcardiogram (VCG). Some of the parameters include amplitudes of maximal positive and maximal negative peaks, area of absolute values, area of positive values, area of negative values, number of samples with 70% higher amplitude than that of the highest peak, amplitude and angle of the QRS vector in a VCG plane. They were measured for all heartbeats annotated as N or PVC in all 48 ECG recordings of the MIT-BIH arrhythmia database. Two reference sets for the Kth nearest-neighbours rule were used-global and local. The classification indices obtained with the global reference set were 75.4% specificity and 80.9% sensitivity. Using the local reference set we increased the specificity to 96.7% and the sensitivity to 96.9%. The achieved specificity and sensitivity are comparable with, and greater than, the results reported in the literature.

Real time detection of ventricular fibrillation and tachycardia

The automatic external defibrillator (AED) is a lifesaving device, which processes and analyses the electrocardiogram (ECG) and delivers a defibrillation shock to terminate ventricular fibrillation or tachycardia above 180 bpm. The built-in algorithm for ECG analysis has to discriminate between shockable and non-shockable rhythms and its accuracy, represented by sensitivity and specificity, is aimed at approaching the maximum values of 100%. An algorithm for VF/VT detection is proposed using a band-pass digital filter with integer coefficients, which is very simple to implement in real-time operation. A branch for wave detection is activated for heart rate measurement and an auxiliary parameter calculation. The method was tested with ECG records from the widely recognized databases of the American Heart Association (AHA) and the Massachusetts Institute of Technology (MIT). A sensitivity of 95.93% and a specificity of 94.38% were obtained.

Comparison of four methods for premature ventricular contraction and normal beat clustering

The learning capacity and the classification ability for normal beats and premature ventricular contractions clustering by four classification methods were compared: neural networks (NN), K-th nearest neighbour rule (Knn), discriminant analysis (DA) and fuzzy logic (FL). Twenty-six morphology feature parameters, which include information of amplitude, area, specific interval durations and measurement of the QRS vector in a VCG plane, were defined. One global and two local learning sets were used. The local classifiers achieved better accuracies because of their good adaptability to the patients, while the capacity of the global classifier to process new records without additional learning was expectedly balanced by lower accuracies. NN assure the best results (high and balanced indices for specificity and sensitivity) using one of the local learning set, while the Knn provides the best results with the other local learning set. Using the global learning set DA and the FL methods perform better than the NN and Knn

Shock advisory tool: Detection of life-threatening cardiac arrhythmias and shock success prediction by means of a common parameter set

Abstract The aim of the present work was to study the possibility of a parameter set to assure both reliable detection of shockable rhythms and adequate shock success prediction. A set of 10 parameters, reflecting the frequency characteristics, the variations, the complexity, the periodicity and the symmetry of the ECG signals was subjected to discriminant analysis. The reliability of the derived parameters to provide an adequate shock advisory decision, which accounts the arrhythmia type (shockable or non-shockable) and the possibility for shock success, was studied. Moreover, the influence of different types of artifacts on the accuracy for shockable and non-shockable rhythms classification was evaluated. The shockable rhythm detection ability was estimated towards the AHA recommendations for reliable automatic external defibrillator algorithm performance, while the accuracy for prediction of the shock outcome was compared with the possibility of other proposed methods to differentiate between ventricular fibrillation episodes amenable and non-amenable to defibrillation. The direct comparison of the shockable rhythm detection results with the AHA recommendations for defined rhythm categories proved the adequacy of the processed ECG features to provide accuracy, which meets the AHA performance goal. Besides this, the proposed parameter set proved its adequacy for shock success prediction and the attained prediction accuracy (above 80%) could be considered as acceptable for possible practical application in automatic external defibrillators. The combination of reliable detection and prediction, as well as the fact that the decision for defibrillation will account not only the rhythm type but also the possibility for successful defibrillation, makes the proposed parameter set a reliable tool for automatic external defibrillator shock-advisory algorithms.

Method for ventricular fibrillation detection in the external electrocardiogram using nonlinear prediction

The automatic external defibrillator is a lifesaving device which processes and analyses the electrocardiogram (ECG) and delivers defibrillation shock when necessary. The accuracy of the built-in algorithm for ECG analysis must be very high, with sensitivity and specificity aimed to approach the maximum values of 100%. An algorithm based on nonlinear prediction of the external ECG signal is proposed. It extracts seven parameters characterizing the prediction possibility of the assessed ECG signal. By means of the K-nearest neighbours rule the diagnostic accuracy of different combinations of these parameters was evaluated. Thus the accuracy obtained was higher than 95% with sensitivity and specificity values depending on the combination of parameters. The method was tested with ECG records from the widely recognized databases of the American Heart Association (AHA) and Massachusetts Institute of Technology (MIT).

Assessment of ECG frequency and morphology parameters for automatic classification of life-threatening cardiac arrhythmias

The reliable recognition and adequate electrical shock therapy of life-threatening cardiac states depend on the electrocardiogram (ECG) descriptors which are used by the defibrillator-embedded automatic arrhythmia analysis algorithms. We propose a method for real-time ECG processing and parameter set extraction using band-pass digital filtration and ECG peak detection. Twelve parameters were derived: (i) seven parameters from the band-pass filter output-six threshold parameters and one peak counter; (ii) five parameters from the ECG peak detection branch, which assess the heart rate, the periodicity and the amplitude/slope symmetry of the ECG peaks. The statistical assessment for more than 36 h of cardiac arrhythmia episodes collected from the public AHA and MIT databases showed that some of the parameters achieved high specificity and sensitivity, but there was no parameter providing 100% separation between non-shockable and shockable rhythms. In order to estimate the influence of the wide variety of cardiac arrhythmias and the different artifacts in real recording conditions, we performed a more detailed study for eight non-shockable and four shockable rhythm categories. The combination of the six top-ranked parameters provided specificity: (i) more than 99% for rhythms with narrow supraventricular complexes, premature ventricular contractions, paced beats and bradycardias; (ii) almost 95% for supraventricular tachycardias; (iii) 91.5% for bundle branch blocks; (iv) 92.2% for slow ventricular tachycardias. The attained sensitivity was above 98% for coarse and fine ventricular fibrillations and 94% for the rapid ventricular tachycardias. The accuracy for the noise contaminated non-shockable and shockable signals exceeded 93%. The proposed parameter set guarantees an accuracy that meets the AHA performance goal for each rhythm category and could be a reliable facility for AED shock-advisory algorithms.

Defibrillation shock success estimation by a set of six parameters derived from the electrocardiogram

It is well known that in some cases defibrillator shocks cannot terminate ventricular fibrillation (VF). Repeated failed shocks often may worsen subsequent response to therapy. This study assesses the ability of six parameters derived from the surface electrocardiogram (ECG) to predict defibrillation shock outcome. Using stepwise discriminant analysis, we obtained several discriminant functions, yielding different combinations of sensitivity and specificity for detection of pre-shock ECG segments corresponding to successful versus unsuccessful shocks. The study was performed consecutively for 3, 4 and 5 s ECG time intervals. The prediction accuracy of 72.3% (61.8% sensitivity and 79.6% specificity) with five parameters and 3 s VF segment analysis prior to defibrillation shock could be considered acceptable for possible practical application in automatic external defibrillators.

Pattern recognition and optimal parameter selection in premature ventricular contraction classification

Analyses of electrocardiographic pattern recognition parameters for premature ventricular contraction (PVC) and Normal (N) beat classification are presented. Twenty-six parameters are defined: 11/spl times/2 for the two ECG leads, 3 for vectorcardiogram (VCG) and width of the complex. Some of them include: amplitudes of maximal positive and negative peaks, area of the absolute values, area of positive and negative values, number of samples with 70% higher amplitude then that of the highest peak, amplitude and angle of the QRS vector in a VCG plane. They are measured for all N and PVC heart beats in the MIT-BIH arrhythmia database. The classification ability of each parameter is tested using discriminant analysis. Considering both leads 7 parameters with highest discriminant power for N and PVC are extracted and a specificity of 96.6% and a sensitivity of 90.5% are obtained. Taking into account relatively all parameters a specificity of 97.3% and a sensitivity of 93.3% are achieved.

Bench study of the accuracy of a commercial AED arrhythmia analysis algorithm in the presence of electromagnetic interferences.

This paper presents a bench study on a commercial automated external defibrillator (AED). The objective was to evaluate the performance of the defibrillation advisory system and its robustness against electromagnetic interferences (EMI) with central frequencies of 16.7, 50 and 60 Hz. The shock advisory system uses two 50 and 60 Hz band-pass filters, an adaptive filter to identify and suppress 16.7 Hz interference, and a software technique for arrhythmia analysis based on morphology and frequency ECG parameters. The testing process includes noise-free ECG strips from the internationally recognized MIT-VFDB ECG database that were superimposed with simulated EMI artifacts and supplied to the shock advisory system embedded in a real AED. Measurements under special consideration of the allowed variation of EMI frequency (15.7-17.4, 47-52, 58-62 Hz) and amplitude (1 and 8 mV) were performed to optimize external validity. The accuracy was reported using the American Heart Association (AHA) recommendations for arrhythmia analysis performance. In the case of artifact-free signals, the AHA performance goals were exceeded for both sensitivity and specificity: 99% for ventricular fibrillation (VF), 98% for rapid ventricular tachycardia (VT), 90% for slow VT, 100% for normal sinus rhythm, 100% for asystole and 99% for other non-shockable rhythms. In the presence of EMI, the specificity for some non-shockable rhythms (NSR, N) may be affected in some specific cases of a low signal-to-noise ratio and extreme frequencies, leading to a drop in the specificity with no more than 7% point. The specificity for asystole and the sensitivity for VF and rapid VT in the presence of any kind of 16.7, 50 or 60 Hz EMI simulated artifact were shown to reach the equivalence of sensitivity required for non-noisy signals. In conclusion, we proved that the shock advisory system working in a real AED operates accurately according to the AHA recommendations without artifacts and in the presence of EMI. The results may be affected for specificity in the case of a low signal-to-noise ratio or in some extreme frequency setting.

Intersubject variability and intrasubject reproducibility of 12-lead ECG metrics: Implications for human verification

BACKGROUND Electrocardiogram (ECG) biometrics is an advanced technology, not yet covered by guidelines on criteria, features and leads for maximal authentication accuracy. OBJECTIVE This study aims to define the minimal set of morphological metrics in 12-lead ECG by optimization towards high reliability and security, and validation in a person verification model across a large population. METHODS A standard 12-lead resting ECG database from 574 non-cardiac patients with two remote recordings (>1year apart) was used. A commercial ECG analysis module (Schiller AG) measured 202 morphological features, including lead-specific amplitudes, durations, ST-metrics, and axes. Coefficient of variation (CV, intersubject variability) and percent-mean-absolute-difference (PMAD, intrasubject reproducibility) defined the optimization (PMAD/CV→min) and restriction (CV<30%) criteria for selection of the most stable and distinctive features. Linear discriminant analysis (LDA) validated the non-redundant feature set for person verification. RESULTS AND CONCLUSIONS Maximal LDA verification sensitivity (85.3%) and specificity (86.4%) were validated for 11 optimal features: R-amplitude (I,II,V1,V2,V3,V5), S-amplitude (V1,V2), Tnegative-amplitude (aVR), and R-duration (aVF,V1).