Automatic Cardiac Rhythm Classification With Concurrent Manual Chest Compressions

Abstract

Electrocardiogram (EKG) based classification of out-of-hospital cardiac arrest (OHCA) rhythms is important to guide treatment and to retrospectively elucidate the effects of therapy on patient response. OHCA rhythms are grouped into five categories: ventricular fibrillation (VF) and tachycardia (VT), asystole (AS), pulseless electrical activity (PEA), and pulse-generating rhythms (PR). Clinically these rhythms are grouped into broader categories like shockable (VF/VT), non-shockable (AS/PEA/PR), or organized (ORG, PEA/PR). OHCA rhythm classification is further complicated because EKGs are corrupted by cardiopulmonary resuscitation (CPR) artifacts. The objective of this study was to demonstrate a framework for automatic multiclass OHCA rhythm classification in the presence of CPR artifacts. In total, 2133 EKG segments from 272 OHCA patients were used: 580 AS, 94 PR, 953 PEA, 479 VF, and 27 VT. CPR artifacts were adaptively filtered, 93 features were computed from the stationary wavelet transform analysis, and random forests were used for classification. A repeated stratified nested cross-validation procedure was used for feature selection, parameter tuning, and model assessment. Data were partitioned patient-wise. The classifiers were evaluated using per class sensitivity, and the unweighted mean of sensitivities (UMS) as a global performance metric. Four levels of clinical detail were studied: shock/no-shock, shock/AS/ORG, VF/VT/AS/ORG, and VF/VT/AS/PEA/PR. The median UMS (interdecile range) for the 2, 3, 4, and 5-class classifiers were: 95.4% (95.1-95.6), 87.6% (87.3-88.1), 80.6% (79.3-81.8), and 71.9% (69.5-74.6), respectively. For shock/no-shock decisions sensitivities were 93.5% (93.0-93.9) and 97.2% (97.0-97.4), meeting clinical standards for artifact-free EKG. The UMS for five classes with CPR artifacts was 5.8-points below that of the best algorithms without CPR artifacts, but improved the UMS of latter by over 19-points for EKG with CPR artifacts. A robust and accurate approach for multiclass OHCA rhythm classification during CPR has been demonstrated, improving the accuracy of the current state-of-the-art methods.INDEX TERMS Out-of-hospital cardiac arrest (OHCA), electrocardiogram (EKG), cardiopulmonary resuscitation (CPR), adaptive filter, stationary wavelet transform (SWT), random forest (RF) classifier. I. INTRODUCTIONOut-of-hospital cardiac arrest (OHCA) is a leading cause of death in the industrialized world. In Europe the estimated annual average incidence of ambulance treated casesThe associate editor coordinating the review of this manuscript and approving it for publication was Nuno Garcia. is 41 (range 19-104) per 100 000 persons [1]. Patients in cardiac arrest lose their cardiac and respiratory function, and die within minutes if not treated. Treatment consists of highly time-sensitive interventions such as: recognition, call for help, cardiopulmonary resuscitation (CPR), defibrillation, and post-resuscitation care. Bystanders and lay rescuers can provide CPR to maintain an artificial perfus...

Rhythm Analysis during Cardiopulmonary Resuscitation Using Convolutional Neural Networks

“…Shockable rhythms comprised lethal ventricular arrhythmia, predominantly VF but also pulseless VT. Non-shockable rhythms included asystole (AS), the absence of electrical activity, and organized rhythms (ORG), or rhythms with visible QRS complexes. The OHCA episodes had median (interquartile range, IQR) durations of 26 min (17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30)(31)(32)(33). From these episodes 15.5 s segments were automatically extracted following these criteria: unique rhythm type in the segment and an interval of 12.5 s with ongoing compressions followed or preceded by a 3 s interval without compressions.…”

“…The latter have shown the highest Se/Sp values by exploiting recent advances in ECG feature extraction and classical machine learning algorithms. ECG features are customarily computed in time, frequency or time-frequency domains [ 23 , 24 , 25 , 26 ]. These features have been efficiently combined using classical machine learning classification algorithms like support vector machines (SVM) or random forests (RF) [ 22 , 25 , 26 ].…”

Multimodal Biosignal Analysis Algorithm for the Classification of Cardiac Rhythms During Resuscitation

“…Second, the addition of the TI signal [8], [15]. And finally, an improved ECG feature extraction based on the SWT [7], [12].…”

“…• ECG features (65 from the ECG and its $ − " coefficients): SampEn, x1, x2, bCP, count1, count2, count3, bWT, Npeak, vfleak, expmod, hilb, IQR, FQR, tcsc, mav, frqbin, cm, kurt, A1, A2, A3, AR_c, AR_n ,m2_s, m3_s, m4_s, v1, v2, v3, v4, v5, v6, v7, v8, v9, Pf0, LAC, Pnh, mSl. A detailed description of the features can be found in [7], [13], [14] and [15]. • TI features (30 from the TI and its $ − % coefficients and its combination with the ECG): the mean power of the TI and the mean cross-power between ECG and TI [16], the first-order derivative of the TI and the peak amplitude of the Fast Fourier Transform (FFT) of the first-order derivative of the TI [17], and the magnitudesquared coherence and the cross power spectral density of ECG and TI.…”

“…Rad et al [6] introduced 5-class OHCA rhythm classification using the ECG. Since then, other ECG-based 5-class classifiers have been developed [7]. However, using only the ECG may limit classification accuracy, particularly the distinction of rhythms that albeit having visible QRS complexes may or may not have a palpable associated mechanical activity.…”