Comparative assessment of shockable ECG rhythm detection algorithms in automated external defibrillators

SHOCKABLE VERSUS NONSHOCKABLE LIFE-THREATENING VENTRICULAR ARRHYTHMIAS USING DWT AND NONLINEAR FEATURES OF ECG SIGNALS

Shockable ventricular arrhythmias (VAs) such as ventricular tachycardia (VT) and ventricular fibrillation (VFib) are the life-threatening conditions requiring immediate attention. Cardiopulmonary resuscitation (CPR) and defibrillation are the significant immediate recommended treatments for these shockable arrhythmias to obtain the return of spontaneous circulation. However, accurate classification of these shockable VAs from nonshockable ones is the key step during defibrillation by automated external defibrillator (AED). Therefore, in this work, we have proposed a novel algorithm for an automated differentiation of shockable and nonshockable VAs from electrocardiogram (ECG) signal. The ECG signals are segmented into 5, 8 and 10[Formula: see text]s. These segmented ECGs are subjected to four levels of discrete wavelet transformation (DWT). Various nonlinear features such as approximate entropy ([Formula: see text], signal energy ([Formula: see text]), Fuzzy entropy ([Formula: see text]), Kolmogorov Sinai entropy ([Formula: see text], permutation entropy ([Formula: see text]), Renyi entropy ([Formula: see text]), sample entropy ([Formula: see text]), Shannon entropy ([Formula: see text]), Tsallis entropy ([Formula: see text]), wavelet entropy ([Formula: see text]), fractal dimension ([Formula: see text]), Kolmogorov complexity ([Formula: see text]), largest Lyapunov exponent ([Formula: see text]), recurrence quantification analysis (RQA) parameters ([Formula: see text]), Hurst exponent ([Formula: see text]), activity entropy ([Formula: see text]), Hjorth complexity ([Formula: see text]), Hjorth mobility ([Formula: see text]), modified multi scale entropy ([Formula: see text]) and higher order statistics (HOS) bispectrum ([Formula: see text]) are obtained from the DWT coefficients. Later, these features are subjected to sequential forward feature selection (SFS) method and selected features are then ranked using seven ranking methods namely, Bhattacharyya distance, entropy, Fuzzy maximum relevancy and minimum redundancy (mRMR), receiver operating characteristic (ROC), Student’s [Formula: see text]-test, Wilcoxon and ReliefF. These ranked features are supplied independently into the [Formula: see text]-Nearest Neighbor (kNN) classifier. Our proposed system achieved maximum accuracy, sensitivity and specificity of (i) 97.72%, 94.79% and 98.74% for 5[Formula: see text]s, (ii) 98.34%, 95.49% and 99.14% for 8[Formula: see text]s and (iii) 98.32%, 95.16% and 99.20% for 10[Formula: see text]s of ECG segments using only ten features. The integration of the proposed algorithm with ECG acquisition systems in the intensive care units (ICUs) can help the clinicians to decipher the shockable and nonshockable life-threatening arrhythmias accurately. Hence, doctors can use the CPR or AED immediately and increase the chance of survival during shockable life-threatening arrhythmia intervals.

A reliable method for rhythm analysis during cardiopulmonary resuscitation

Interruptions in cardiopulmonary resuscitation (CPR) compromise defibrillation success. However, CPR must be interrupted to analyze the rhythm because although current methods for rhythm analysis during CPR have high sensitivity for shockable rhythms, the specificity for nonshockable rhythms is still too low. This paper introduces a new approach to rhythm analysis during CPR that combines two strategies: a state-of-the-art CPR artifact suppression filter and a shock advice algorithm (SAA) designed to optimally classify the filtered signal. Emphasis is on designing an algorithm with high specificity. The SAA includes a detector for low electrical activity rhythms to increase the specificity, and a shock/no-shock decision algorithm based on a support vector machine classifier using slope and frequency features. For this study, 1185 shockable and 6482 nonshockable 9-s segments corrupted by CPR artifacts were obtained from 247 patients suffering out-of-hospital cardiac arrest. The segments were split into a training and a test set. For the test set, the sensitivity and specificity for rhythm analysis during CPR were 91.0% and 96.6%, respectively. This new approach shows an important increase in specificity without compromising the sensitivity when compared to previous studies.

ECG spectral and morphological parameters reviewed and updated to detect adult and paediatric life-threatening arrhythmia

Since the International Liaison Committee on Resuscitation approved the use of automated external defibrillators (AEDs) in children, efforts have been made to adapt AED algorithms designed for adult patients to detect paediatric ventricular arrhythmias accurately. In this study, we assess the performance of two spectral (A(2) and VFleak) and two morphological parameters (TCI and CM) for the detection of lethal ventricular arrhythmias using an American Heart Association (AHA) compliant database that includes adult and paediatric arrhythmias. Our objective was to evaluate how those parameters can be optimally adjusted to discriminate shockable from nonshockable rhythms in adult and paediatric patients. A total of 1473 records were analysed: 751 from 387 paediatric patients (<or=16 years of age) and 722 records from 381 adult patients. The spectral parameters showed no significant differences (p > 0.01) between the adult and paediatric patients for the shockable records; the differences for nonshockable records however were significant. Still, these parameters maintained the discrimination power when paediatric rhythms were included. A single threshold could be adjusted to obtain sensitivities and specificities above the AHA goals for the complete database. The sensitivities for ventricular fibrillation (VF) and ventricular tachycardia (VT) were 91.1% and 96.6% for VFleak, and 90.3% and 99.3% for A(2). The specificities for normal sinus rhythm (NSR) and other nonshockable rhythms were 99.5% and 96.3% for VFleak, and 99.0% and 97.7% for A(2). On the other hand, the morphological parameters showed significant differences between the adult and paediatric patients, particularly for the nonshockable records, because of the faster heart rates of the paediatric rhythms. Their performance clearly degraded with paediatric rhythms. Using a single threshold, the sensitivities and specificities were below the AHA goals, particularly VT sensitivity (60.4% for TCI and 65.8% for CM) and the specificity for other nonshockable rhythms (51.7% for TCI and 34.5% for CM). The specificities, particularly for the adult case, improve when the thresholds are independently adjusted for each adult and paediatric database.

Identifying potentially shockable rhythms without interrupting cardiopulmonary resuscitation

OBJECTIVE

Current versions of automated external defibrillators (AEDs) mandate interruptions of chest compression for rhythm analyses because of artifacts produced by chest compressions. Interruption of chest compressions reduces likelihood of successful resuscitation by as much as 50%. We sought a method to identify a shockable rhythm without interrupting chest compressions during cardiopulmonary resuscitation (CPR).

DESIGN

Experimental study.

SETTING

Weil Institute of Critical Care Medicine, Rancho Mirage, CA.

SUBJECTS

None.

INTERVENTIONS

Electrocardiographs (ECGs) were recorded in conjunction with AEDs during CPR in human victims. A shockable rhythm was defined as disorganized rhythm with an amplitude > 0.1 mV or, if organized, at a rate of > or = 180 beats/min. Wavelet-based transformation and shape-based morphology detection were used for rhythm classification. Morphologic consistencies of waveform representing QRS components were analyzed to differentiate between disorganized and organized rhythms. For disorganized rhythms, the amplitude spectrum area was computed in the frequency domain to distinguish between shockable ventricular fibrillation and nonshockable asystole. For organized rhythms, in victims in whom the absence of a heartbeat was independently confirmed, the heart rate was estimated for further classification.

MEASUREMENTS AND MAIN RESULTS

To derive the algorithm, we used 29 recordings on 29 patients from the Creighton University ventricular tachyarrhythmia database. For validation, the algorithm was tested on an independent population of 229 victims, including recordings of both ECG and depth of chest compressions obtained during suspected out-of-hospital sudden death. The recordings included 111 instances in which the ECG was corrupted during chest compressions. A shockable rhythm was identified with a sensitivity of 93% and a specificity of 89%, yielding a positive predictive value of 91%. A nonshockable rhythm was identified with a sensitivity of 89%, a specificity of 93%, and a positive predictive value of 91% during uninterrupted chest compression.

CONCLUSIONS

The algorithm fulfilled the potential lifesaving advantages of allowing for uninterrupted chest compression, avoiding pauses for automated rhythm analyses before prompting delivery of an electrical shock.

Automated detection of ventricular fibrillation to guide cardiopulmonary resuscitation

Sudden death due to ventricular fibrillation (VF) is a catastrophic event, especially in out-of-hospital settings. Prompt detection of VF and preparedness to intervene with cardiopulmonary resuscitation (CPR) and especially the delivery of an electrical shock is potentially lifesaving. The reliability and accuracy of automated VF detection by current versions of automated external defibrillators (AEDs) require interruption of CPR because the ECG signal, which is the source of rhythm detection, is corrupted by chest compressions. Significantly better outcomes have been reported if effective chest compression precedes electrical defibrillation and especially if interruptions are minimized. We therefore sought a method by which VF detection could proceed without interrupting chest compressions. A VF detection algorithm was therefore derived based on a method by which continuous wavelet transform is used, together with measurement of morphologic consistency. This method was intended to distinguish between disorganized and organized rhythms. The Fourier-transform-based amplitude spectrum analysis was then used to detect the likelihood that VF was the rhythm prompting the delivery of an electrical shock. The algorithm was validated on 33,095 electrocardiographic segments, including 8840 segments corrupted by compression artifacts from 232 patients after out-of-hospital cardiac arrest. Nine thousand one hundred eighty-seven of 10,042 VF segments and 20,884 of 23,053 non-VF segments were correctly classified, with a sensitivity of 91.5% and a specificity of 90.6%. Although the proposed algorithm has a lesser predictive value for VF detection than the uncorrupted ECGs in clinical settings, it has the major potential for automated rhythm identification to guide defibrillation without repetitive interruptions of CPR.

Automated external defibrillation in cardiac surgery

Revision open heart surgery may be impeded by a dense network of pericardial adhesions rendering cardiac mobilization laborious or incomplete, and internal defibrillation impossible. External defibrillation, the current alternative to internal defibrillation, may result in myocardial stunning secondary to the delivery of escalating, monophasic, high-energy shocks. Automated external defibrillation, by delivering consecutive, non-escalating, impedance-compensated, low-energy, biphasic electric shocks to the myocardium, may provide a more effective and safer option whilst reducing the risk of myocardial stunning.

Automatic external defibrillator: key link in the chain of survival

INTRODUCTION

Sudden cardiac death is a major health problem. Worldwide success of resuscitation from out-of-hospital cardiac arrest is modest, with 5% to 10% survival to hospital discharge.

METHODS AND RESULTS

In the chain of survival, early defibrillation (goal <5 min after collapse) is a major determinant of successful outcome of resuscitation. This goal is rarely achieved, but the automatic external defibrillator (AED) is a promising tool for lay defibrillation. The AED is a safe and effective device with nearly 100% accurate detection of ventricular fibrillation and nearly 100% accurate detection of a nonshockable rhythm. A large uncontrolled experience suggests improved outcome in nontraditional responders such as police. Controlled studies of community application of the AED are under way.

CONCLUSION

The AED is a promising tool in the fight against sudden cardiac death and should be studied and supported by all scientists involved, including electrophysiologists.

Physician interpretation and quantitative measures of electrocardiographic ventricular fibrillation waveform

OBJECTIVES

The characteristics of the ventricular fibrillation (VF) waveform may influence treatment decisions and the likelihood of therapeutic success. However, assessment of VF as being fine or coarse and the distinction between fine VF and asystole are largely subjective. The authors sought to determine the level of agreement among physicians for interpretation of varying VF waveforms, and to compare these subjective interpretations with quantitative measures.

METHODS

Six-second segments of waveform from LIFEPAK 300 units were collected. Fifty segments, including 45 VF and five ventricular tachycardia (VT) distracters, were graphed to simulate rhythm strips. These waveforms were quantitatively described using scaling exponent, root-mean-squared amplitude, and centroid frequency. Thirty-two emergency medicine residents were asked to interpret the arrhythmias as VT, "coarse" VF, "fine" VF, or asystole. Their responses were compared with the qantitative measures. Interphysician agreement was assessed with the kappa statistic.

RESULTS

One thousand four hundred forty interpretations were analyzed. There was fair agreement between physicians about the classification of arrhythmias (kappa = 0.39). Mean values associated with coarse VF, fine VF, and asystole differed in all three quantitative measure categories. The decision whether to defibrillate was highly correlated with the distinction between VF and asystole (Pearson chi-square = 1,170.40, df = 1, p[two-sided] < 0.001).

CONCLUSIONS

With only fair agreement on the threshold of fine VF and asystole, defibrillation decisions are largely subjective and caregiver-specific. These data suggest that quantitative measures of the VF waveform could augment the current standard of subjective classification of VF by emergency care providers.