Cardiopulmonary Resuscitation Artifact Research Articles

Transfer learning based deep network for signal restoration and rhythm analysis during cardiopulmonary resuscitation using only the ECG waveform

Minimizing the interruption of cardiopulmonary resuscitation (CPR) is an important technique to improve the survival of out-of-hospital cardiac arrest (OHCA) patients. Recent studies have adopted deep convolutional neural networks (CNNs) to analyze corrupted ECGs during CPR for shockable rhythm detection, however, none of them provided restored ECG signals using only the ECG waveform. In this study, a self-supervised UNet deep learning network is designed to restore the underlying ECG signals during chest compressions without additional reference signals. The UNet is pretrained using a large dataset created by combining 12-lead clean ECG signals with simulated CPR artifacts generated by a mathematical model. Transfer learning is applied to fine-tune the UNet parameters based on combined ECG signals from OHCA patients. Finally, an independent dataset extracted from OHCA patients is used to test the model. Compared with the artifact corrupted ECG signals, the overall signal-to-noise ratio (SNR) is improved from −5.3 (-11.2, 0.8) dB to 1.9 (0.7, 3.9) dB after the restoration, and the accuracy of the rhythm detection is improved from 65.8% to 90.8%. The proposed method not only effectively restores the underlying ECG signals but also provides accurate shock advice without interrupting CPR using only the ECG.

Enhancing the accuracy of shock advisory algorithms in automated external defibrillators during ongoing cardiopulmonary resuscitation using a deep convolutional Encoder-Decoder filtering model

Survival from out-of-hospital cardiac arrests (OHCA) depends on an accurate defibrillatory shock decision during cardiopulmonary resuscitation (CPR). Since chest compressions induce severe motion artifact in the electrocardiogram (ECG), current automatic external defibrillators (AEDs) instruct the user not to perform CPR during the rhythm analysis period. However, performing continuous CPR is vital and dramatically increases the chance of survival. Hence, we demonstrate a novel application of a deep convolutional neural network encoder-decoder (CNNED) method to suppress CPR artifact in near real-time, using only ECG data. The encoder portion of the CNNED uses the magnitude and phase contents derived via time-varying spectral analysis to learn distinct features that are representative of both the ECG signal and CPR artifact. The decoder portion takes the results from the encoder and reconstructs what is perceived as the motion artifact-removed ECG data. These procedures are done via a multitude of training of the CNNED using many different arrhythmias contaminated with CPR. In this study, CPR-contaminated ECGs were generated by combining clean ECGs with 52 different CPR artifacts. ECG data from CUDB, VFDB, SDDB, and AFDB datasets which belong to the Physionet’s Physiobank archive were used to create the training set containing 14-second ECG segments. The performance of the proposed CNNED was evaluated on a separate test set comprised of 23,816 CPR-contaminated 14-second ECG segments from 458 subjects. The results were evaluated by three metrics: signal-to-noise ratio (SNR), correlation coefficient, and accuracy of Defibtech’s AED shock advisory algorithm (SAA). CNNED resulted in an increase of the mean SNR value from −3 dB to 7.5 dB and 7.1 dB for shockable and non-shockable rhythms, respectively. 80.85% of the filtered shockable and 74.13% of the filtered non-shockable ECG data were highly correlated (>0.7) with the artifact-free ECG; these values were only 13.30% and 12.71% for CPR-contaminated shockable and non-shockable, respectively, without our filtering approach. Comparing results of Defibtech’s AED SAA before and after applying CNNED on the CPR-contaminated ECG, the specificity improved from 96.21% to 99.14% for normal sinus rhythm, and from 88.5% to 96.45% for other non-shockable rhythms. The sensitivity of shockable detection also increased from 67.68% to 90.90% for ventricular fibrillation, and from 62.71% to 82.26% for ventricular tachycardia. These results indicate continuous and accurate AED rhythm analysis without stoppage of CPR, using only ECG data.

Automated Condition-Based Suppression of the CPR Artifact in ECG Data to Make a Reliable Shock Decision for AEDs during CPR.

Cardiopulmonary resuscitation (CPR) corrupts the morphology of the electrocardiogram (ECG) signal, resulting in an inaccurate automated external defibrillator (AED) rhythm analysis. Consequently, most current AEDs prohibit CPR during the rhythm analysis period, thereby decreasing the survival rate. To overcome this limitation, we designed a condition-based filtering algorithm that consists of three stop-band filters which are turned either ‘on’ or ‘off’ depending on the ECG’s spectral characteristics. Typically, removing the artifact’s higher frequency peaks in addition to the highest frequency peak eliminates most of the ECG’s morphological disturbance on the non-shockable rhythms. However, the shockable rhythms usually have dynamics in the frequency range of (3–6) Hz, which in certain cases coincide with CPR compression’s harmonic frequencies, hence, removing them may lead to destruction of the shockable signal’s dynamics. The proposed algorithm achieves CPR artifact removal without compromising the integrity of the shockable rhythm by considering three different spectral factors. The dataset from the PhysioNet archive was used to develop this condition-based approach. To quantify the performance of the approach on a separate dataset, three performance metrics were computed: the correlation coefficient, signal-to-noise ratio (SNR), and accuracy of Defibtech’s shock decision algorithm. This dataset, containing 14 s ECG segments of different types of rhythms from 458 subjects, belongs to Defibtech commercial AED’s validation set. The CPR artifact data from 52 different resuscitators were added to artifact-free ECG data to create 23,816 CPR-contaminated data segments. From this, 82% of the filtered shockable and 70% of the filtered non-shockable ECG data were highly correlated (>0.7) with the artifact-free ECG; this value was only 13 and 12% for CPR-contaminated shockable and non-shockable, respectively, without our filtering approach. The SNR improvement was 4.5 ± 2.5 dB, averaging over the entire dataset. Defibtech’s rhythm analysis algorithm was applied to the filtered data. We found a sensitivity improvement from 67.7 to 91.3% and 62.7 to 78% for VF and rapid VT, respectively, and specificity improved from 96.2 to 96.5% and 91.5 to 92.7% for normal sinus rhythm (NSR) and other non-shockables, respectively.

Abstract 12792: Enhancing the Accuracy of Cardiac Rhythm Analysis in Automated External Defibrillators During Ongoing Cardiopulmonary Resuscitation by Applying a Deep Encoder-Decoder Filtering Model

Survival from out-of-hospital cardiac arrests depends on an accurate defibrillatory shock decision during cardiopulmonary resuscitation (CPR). Since chest compressions induce severe motion artifact in the electrocardiogram (ECG), current automatic external defibrillators (AEDs) do not perform CPR during the rhythm analysis period. However, performing continuous CPR is vital and dramatically increases the chance of survival. Hence, we demonstrate a novel application of a deep convolutional neural network encoder-decoder (CNNED) method in suppressing CPR artifact in near real-time using only ECG data. The encoder portion of the CNNED uses the frequency and phase contents derived via time-varying spectral analysis to learn distinct features that are representative of both the ECG signal and CPR artifact. The decoder portion takes the results from the encoder and reconstructs what is perceived as the motion artifact removed ECG data. These procedures are done via multitude of training of CNNED using many different arrhythmia contaminated with CPR. In this study, CPR-contaminated ECGs were generated by combining clean ECG with CPR artifacts from 52 different performers. ECG data from CUDB, VFDB, and SDDB datasets which belong to the Physionet’s Physiobank archive were used to create the training set containing 89,984 14-second ECG segments. The performance of the proposed CNNED was evaluated on a separate test set comprising of 23,816 CPR-contaminated 14-second ECG segments from 458 subjects. The results were evaluated by two metrics: signal-to-noise ratio (SNR), and Defibtech’s AED rhythm analysis algorithm. CNNED resulted in the increase of the mean SNR value from -3 dB to 5.63 dB and 6.3 dB for shockable and non-shockable rhythms, respectively. Comparing Defibtech’s AED rhythm classifier before and after applying CNNED on the CPR-contaminated ECG, the specificity improved from 96.57% to 99.31% for normal sinus rhythm, and from 91.5% to 96.57% for other non-shockable rhythms. The sensitivity of shockable detection also increased from 67.68% to 87.76% for ventricular fibrillation, and from 62.71% to 81.27% for ventricular tachycardia. These results indicate continuous and accurate AED rhythm analysis without stoppage of CPR using only ECG data.

Optimization of End-to-End Convolutional Neural Networks for Analysis of Out-of-Hospital Cardiac Arrest Rhythms during Cardiopulmonary Resuscitation.

High performance of the shock advisory analysis of the electrocardiogram (ECG) during cardiopulmonary resuscitation (CPR) in out-of-hospital cardiac arrest (OHCA) is important for better management of the resuscitation protocol. It should provide fewer interruptions of chest compressions (CC) for non-shockable organized rhythms (OR) and Asystole, or prompt CC stopping for early treatment of shockable ventricular fibrillation (VF). Major disturbing factors are strong CC artifacts corrupting raw ECG, which we aimed to analyze with optimized end-to-end convolutional neural network (CNN) without pre-filtering or additional sensors. The hyperparameter random search of 1500 CNN models with 2–7 convolutional layers, 5–50 filters and 5–100 kernel sizes was done on large databases from independent OHCA interventions for training (3001 samples) and validation (2528 samples). The best model, named CNN3-CC-ECG network with three convolutional layers (filters@kernels: 5@5,25@20,50@20) presented Sensitivity Se(VF) = 89%(268/301), Specificity Sp(OR) = 91.7%(1504/1640), Sp(Asystole) = 91.1%(3325/3650) on an independent test OHCA database. CNN3-CC-ECG’s ability to effectively extract features from raw ECG signals during CPR was comprehensively demonstrated, and the dependency on the CPR corruption level in ECG was tested. We denoted a significant drop of Se(VF) = 74.2% and Sp(OR) = 84.6% in very strong CPR artifacts with a signal-to-noise ratio of SNR < −9 dB, p < 0.05. Otherwise, for strong, moderate and weak CC artifacts (SNR > −9 dB, −6 dB, −3 dB), we observed insignificant performance differences: Se(VF) = 92.5–96.3%, Sp(OR) = 93.4–95.5%, Sp(Asystole) = 92.6–94.0%, p > 0.05. Performance stability with respect to CC rate was validated. Generalizable application of the optimized computationally efficient CNN model was justified by an independent OHCA database, which to our knowledge is the largest test dataset with real-life cardiac arrest rhythms during CPR.

Shock decision algorithm for use during load distributing band cardiopulmonary resuscitation

AimChest compressions delivered by a load distributing band (LDB) induce artefacts in the electrocardiogram. These artefacts alter shock decisions in defibrillators. The aim of this study was to demonstrate the first reliable shock decision algorithm during LDB compressions. MethodsThe study dataset comprised 5813 electrocardiogram segments from 896 cardiac arrest patients during LDB compressions. Electrocardiogram segments were annotated by consensus as shockable (1154, 303 patients) or nonshockable (4659, 841 patients). Segments during asystole were used to characterize the LDB artefact and to compare its characteristics to those of manual artefacts from other datasets. LDB artefacts were removed using adaptive filters. A machine learning algorithm was designed for the shock decision after filtering, and its performance was compared to that of a commercial defibrillator's algorithm. ResultsMedian (90% confidence interval) compression frequencies were lower and more stable for the LDB than for the manual artefact, 80 min−1 (79.9–82.9) vs. 104.4 min−1 (48.5–114.0). The amplitude and waveform regularity (Pearson's correlation coefficient) were larger for the LDB artefact, with 5.5 mV (0.8–23.4) vs. 0.5 mV (0.1–2.2) (p < 0.001) and 0.99 (0.78–1.0) vs. 0.88 (0.55–0.98) (p < 0.001). The shock decision accuracy was significantly higher for the machine learning algorithm than for the defibrillator algorithm, with sensitivity/specificity pairs of 92.1/96.8% (machine learning) vs. 91.4/87.1% (defibrillator) (p < 0.001). ConclusionCompared to other cardiopulmonary resuscitation artefacts, removing the LDB artefact was challenging due to larger amplitudes and lower compression frequencies. The machine learning algorithm achieved clinically reliable shock decisions during LDB compressions.

Deep Neural Network Approach for Continuous ECG-Based Automated External Defibrillator Shock Advisory System During Cardiopulmonary Resuscitation.

BackgroundBecause chest compressions induce artifacts in the ECG, current automated external defibrillators instruct the user to stop cardiopulmonary resuscitation (CPR) while an automated rhythm analysis is performed. It has been shown that minimizing interruptions in CPR increases the chance of survival.Methods and ResultsThe objective of this study was to apply a deep‐learning algorithm using convolutional layers, residual networks, and bidirectional long short‐term memory method to classify shockable versus nonshockable rhythms in the presence and absence of CPR artifact. Forty subjects' data from Physionet with 1131 shockable and 2741 nonshockable samples contaminated with 43 different CPR artifacts that were acquired from a commercial automated external defibrillator during asystole were used. We had separate data as train and test sets. Using our deep neural network model, the sensitivity and specificity of the shock versus no‐shock decision for the entire data set over the 4‐fold cross‐validation sets were 95.21% and 86.03%, respectively. This result was based on the training and testing of the model using ECG data in both the presence and the absence of CPR artifact. For ECG without CPR artifact, the sensitivity was 99.04% and the specificity was 95.2%. A sensitivity of 94.21% and a specificity of 86.14% were obtained for ECG with CPR artifact. In addition to 4‐fold cross‐validation sets, we also examined leave‐one‐subject‐out validation. The sensitivity and specificity for the case of leave‐one‐subject‐out validation were 92.71% and 97.6%, respectively.ConclusionsThe proposed trained model can make shock versus nonshock decision in automated external defibrillators, regardless of CPR status. The results meet the American Heart Association's sensitivity requirement (>90%).

A method to predict ventricular fibrillation shock outcome during chest compressions

BackgroundOut-of-hospital ventricular fibrillation (VF) cardiac arrest is a leading cause of death. Quantitative analysis of the VF electrocardiogram (ECG) can predict patient outcomes and could potentially enable a patient-specific, guided approach to resuscitation. However, VF analysis during resuscitation is confounded by cardiopulmonary resuscitation (CPR) artifact in the ECG, challenging continuous application to guide therapy throughout resuscitation. We therefore sought to design a method to predict VF shock outcomes during CPR. MethodsStudy data included 4577 5-s VF segments collected during and without CPR prior to defibrillation attempts in N = 1151 arrest patients. Using training data (460 patients), an algorithm was designed to predict the VF shock outcomes of defibrillation success (return of organized ventricular rhythm) and functional survival (Cerebral Performance Category 1–2). The algorithm was designed with variable-frequency notch filters to reduce CPR artifact in the ECG based on real-time chest compression rate. Ten ECG features and three dichotomous patient characteristics were developed to predict outcomes. These variables were combined using support vector machines and logistic regression. Algorithm performance was evaluated by area under the receiver operating characteristic curve (AUC) to predict outcomes in validation data (691 patients). ResultsAUC (95% Confidence Interval) for predicting defibrillation success was 0.74 (0.71–0.77) during CPR and 0.77 (0.74–0.79) without CPR. AUC for predicting functional survival was 0.75 (0.72–0.78) during CPR and 0.76 (0.74–0.79) without CPR. ConclusionA novel algorithm predicted defibrillation success and functional survival during ongoing CPR following VF arrest, providing a potential proof-of-concept towards real-time guidance of resuscitation therapy.

Abstract 238: A Deep Neural Network Approach for Continuous Electrocardiogram-based Shock Advisory System During Cardio Pulmonary Resuscitation

The ability of an automatic external defibrillator (AED) to make a reliable shock decision during cardio pulmonary resuscitation (CPR) would improve the survival rate of patients with out-of-hospital cardiac arrest. Since chest compressions induce motion artifacts in the electrocardiogram (ECG), current AEDs instruct the user to stop CPR while an automated rhythm analysis is performed. It has been shown that minimizing interruptions in CPR increases the chance of survival. While deep learning approaches have been used successfully for arrhythmia classification, their performance has not been evaluated for creating an AED shock advisory system that can coexist with CPR. To this end, the objective of this study was to apply a deep-learning algorithm using convolutional layers and residual networks to classify shockable versus non-shockable rhythms in the presence and absence of CPR artifact using only the ECG data. The feasibility of the deep learning method was validated using 8-sec segments of ECG with and without CPR. Two separate databases were used: 1) 40 subjects’ data without CPR from Physionet with 1131 shockable and 2741 non-shockable classified recordings, and 2) CPR artifacts that were acquired from a commercial AED during asystole delivered by 43 different resuscitators. For each 8-second ECG segment, randomly chosen CPR data from 43 different types were added to it so that 5 non-shockable and 10 shockable CPR-contaminated ECG segments were created. We used 30 subjects’ and the remaining 10 for training and test datasets, respectively, for the database 1). For the database 2), we used 33 and 10 subjects’ data for training and testing, respectively. Using our deep neural network model, the sensitivity and specificity of the shock versus no-shock decision for both datasets using the four-fold cross-validation were found to be 95.21% and 86.03%, respectively. For shockable versus non-shockable classification of ECG without CPR artifact, the sensitivity was 99.04% and the specificity was 95.2%. A sensitivity of 94.21% and a specificity of 86.14% were obtained for ECG with CPR artifact. These results meet the AHA sensitivity requirement (&gt;90%).

Restoration of the electrocardiogram during mechanical cardiopulmonary resuscitation

Objective: An artefact-free electrocardiogram (ECG) is essential during cardiac arrest to decide therapy such as defibrillation. Mechanical cardiopulmonary resuscitation (CPR) devices cause movement artefacts that alter the ECG. This study analyzes the effectiveness of mechanical CPR artefact suppression filters to restore clinically relevant ECG information. Approach: In total, 495 10 s ECGs were used, of which 165 were in ventricular fibrillation (VF), 165 in organized rhythms (OR) and 165 contained mechanical CPR artefacts recorded during asystole. CPR artefacts and rhythms were mixed at controlled signal-to-noise ratios (SNRs), ranging from –20 dB to 10 dB. Mechanical artefacts were removed using least mean squares (LMS), recursive least squares (RLS) and Kalman filters. Performance was evaluated by comparing the clean and the restored ECGs in terms of restored SNR, correlation-based similarity measures, and clinically relevant features: QRS detection performance for OR, and dominant frequency, mean amplitude and waveform irregularity for VF. For each filter, a shock/no-shock support vector machine algorithm based on multiresolution analysis of the restored ECG was designed, and evaluated in terms of sensitivity (Se) and specificity (Sp). Main results: The RLS filter produced the largest correlation coefficient (0.80), the largest average increase in SNR (9.5 dB), and the best QRS detection performance. The LMS filter best restored VF with errors of 10.3% in dominant frequency, 18.1% in amplitude and 11.8% in waveform irregularity. The Se/Sp of the diagnosis of the restored ECG were 95.1/94.5% using the RLS filter and 97.0/91.4% using the LMS filter. Significance: Suitable filter configurations to restore ECG waveforms during mechanical CPR have been determined, allowing reliable clinical decisions without interrupting mechanical CPR therapy.

Abstract 237: Using the Thoracic Impedance to Predict Measures From Invasive Arterial Blood Pressure in Out-Of-Hospital Cardiac Arrest

Background: During cardiopulmonary resuscitation (CPR), pulse detection can be challenging. Invasive blood pressure measurements (IBP) can help monitoring patient hemodynamics, but arterial catheter placement is difficult. Transthoracic impedance (TI) measured between the defibrillator pads can detect circulation activity. We hypothesized that TI changes can predict the corresponding IBP, and potentially be used to non-invasively detect pulse during CPR. Materials and methods: We included 28 out of hospital cardiac arrest patients receiving CPR by the Oslo Emergency Service who had concurrent recordings of IBP (radial artery, BD, 20G, US) and TI (via defibrillator pads, LP15, Stryker, US). 5-second segments with stable and CPR artefact free signals were extracted (Figure). The circulation component of the TI signal (Figure, red line) was extracted using a Kalman smoother. Ten waveform features were computed per segment and fed into a random forest regressor to predict systolic and diastolic arterial pressures (SAP, DAP), their difference (DifAP) and area of the IBP signal (ArAP). Pearson correlation coefficients between the regression model and the IBP metrics were computed. Data were divided by patient into training/test sets to fit and evaluate the model, respectively, and the process was repeated 500 times. Results: 235 minutes (2261 segments) were extracted with median (Q1-Q3) values of 71.3(39.2-88.1) mmHg for SAP, 44.2(30.0-50.0) mmHg for DAP, 25.6(7.1-38.8) mmHg for DifAP and 63.4(17.0-85.9) mmHg*sec for ArAP. The correlation coefficients between TI-predicted and IBP-measured SAP, DAP, DifAP and ArAP were 0.62 (0.49-0.72), 0.36 (0.22-0.49), 0.69 (0.57-0.76) and 0.64 (0.50-0.73), respectively. Conclusions: Different hemodynamic phases can be observed in both TI and IBP (Figure). TI-based predictions showed good correlation with IBP measures. This could lead to new non-invasive methods to monitor different phases of circulation based on the TI.

A Robust Machine Learning Architecture for a Reliable ECG Rhythm Analysis during CPR.

Chest compressions delivered during cardiopulmonary resuscitation (CPR) induce artifacts in the ECG that may make the shock advice algorithms (SAA) of defibrillators inaccurate. There is evidence that methods consisting of adaptive filters that remove the CPR artifact followed by machine learning (ML) based algorithms are able to make reliable shock/no-shock decisions during compressions. However, there is room for improvement in the performance of these methods. The objective was to design a robust ML framework for a reliable shock/no-shock decision during CPR. The study dataset contained 596 shockable and 1697 nonshockable ECG segments obtained from 273 cases of out-of-hospital cardiac arrest. Shock/no-shock labels were adjudicated by expert reviewers using ECG intervals without artifacts. First, CPR artifacts were removed from the ECG using a Least Mean Squares (LMS) filter. Then, 38 shock/no-shock decision features based on the Stationary Wavelet Transform (SWT) were extracted from the filtered ECG. A wapper-based feature selection method was applied to select the 6 best features for classification. Finally, 4 state-of-the-art ML classifiers were tested to make the shock/no-shock decision. These diagnoses were compared with the rhythm annotations to compute the Sensitivity (Se) and Specificity (Sp). All classifiers achieved an Se above 94.5%, Sp above 95.5% and an accuracy around 96.0%. They all exceeded the 90% Se and 95% Sp minimum values recommended by the American Heart Association.

A VARIATIONAL MODE DECOMPOSITION BASED APPROACH FOR CARDIOPULMONARY RESUSCITATION ESTIMATION AND ASSESSMENT

Cardiac arrests remain leading causes of deaths for thousands of people annually. One of the most common methods for cardiac arrest treatment is cardiopulmonary resuscitation (CPR) that provides chest compressions. It has been shown that the quality of chest compression is considered as one of key indicators for assessment of CPR performance. In this paper, we present an approach for CPR quality evaluation using ECG contaminated with CPR artifact and thoracic impedance. The proposed approach contains two key steps: First, the CPR artifact signal is estimated via variational mode decomposition (VMD) and a mode selection algorithm based on mode’s frequency and frequency of thoracic impedance signal. In the second step, CPR parameters performed on the estimated CPR signals are computed and compared with those derived by reference CPR. The proposed approach is applied for a dataset including patients presenting with asystole, ventricular tachycardia, and pulseless electrical activity. Quantitative results validate the performance of the proposed approach for CPR quality assessment.

Automatic Cardiac Rhythm Classification With Concurrent Manual Chest Compressions

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.

Evaluation of chest compression artefact removal based on rhythm assessments made by clinicians

AimTo evaluate the performance of a state-of-the-art cardiopulmonary resuscitation (CPR) artefact suppression method by assessing to what extent the filtered electrocardiogram (ECG) can be correctly diagnosed by emergency medicine doctors. MethodsA total of 819 ECG segments were used. Each segment contained two consecutive 10 s intervals, an artefact free interval and an interval corrupted by CPR artefacts. Each ECG segment was digitally processed to remove CPR artefacts using an adaptive filter. Each ECG segment was split into artefact-free and filtered intervals, randomly reordered for dissociation, and independently offered to four reviewers for rhythm annotation. The rhythm annotations of the artefact-free intervals were considered as the gold standard against which the rhythm annotations of the filtered intervals were evaluated. For the filtered intervals, the rater agreement (κ, Kappa score) with the gold standard, the sensitivity and the specificity were computed individually for each reviewer, and jointly through the majority decision of the pool of reviewers (DPR). These results were also compared to those obtained using a commercial shock advisory algorithm (SAA). ResultsThe agreement between each reviewer and the gold standard was moderate ranging between κ = 0.41–0.64. The sensitivities and specificities ranged between 64.3–95.0%, and 70.0–95.9%, respectively. The agreement for the DPR was substantial with κ = 0.64 (0.62–0.66), a sensitivity of 90.6%, and a specificity of 85.6%. For the SAA, the agreement was fair with κ = 0.33 (0.31–0.35), a sensitivity of 90.3%, and a specificity of 66.4%. ConclusionClinicians outperformed the SAA, but specificities remained below the specifications recommended by the American Heart Association. Visual assessment of the filtered ECG by clinicians is not reliable enough, and varies greatly among clinicians. Results considerably improve by considering the consensus decision of a pool of clinicians.

Rhythm analysis in CPR

It's necessary to interrupt cardiopulmonary resuscitation (CPR) for a reliable automatic external defibrillator (AED) rhythm analysis, because the mechanical activity from the chest compressions introduces artifacts in the electrocardiogram (ECG) that substantially lower the capacity of an AED to judge cardio-electric rhythm. However, repeated interruptions of compression will reduce the quality of CPR, which in turn affect the prognosis of patients with cardiac arrest (CA). In order to improve the quality of CPR, reduce the interruptions of chest compression and implement accurate defibrillation, people have made many efforts on identifying ECG rhythm in CPR. The studies can be grouped into two broad categories: those based on the artificial mixture of ECG data and CPR artifacts and those based on CA data recorded during CPR. This article introduced researches for rhythm recognition in CPR, including sources and characteristics of CPR artifacts, methods of rhythm analysis, and provided a basis for the study of how to improve the accuracy of cardio-electric rhythm recognition.

LMS 적응필터를 이용한 직교 함수 기반의CPR 잡음 제거 알고리즘 설계

This study proposes an algorithm for removal of CPR artifact in order that automated external defibrillator (AED) can effectively diagnose ECG rhythm during cardiopulmonary resuscitation (CPR). Current AED required to interrupt chest compression for reliable rhythm analysis to avoid the effect of artifacts produced by CPR. However even temporarily interruption of chest compression during CPR adversely affects the probability of restoration of spontaneous circulation (ROSC) and survival after the delivery of the shock. Therefore, we proposed a method for removal of CPR artifacts using least mean square (LMS) filter. The removal of the CPR artifacts would enable compressions to continue during AED rhythm analysis, thereby increasing the likelihood of resuscitation success. It was tested on 31 segments of shockable and 300 segments of non-shockable ECG signals recorded from three pigs during CPR. In the result, sensitivity (Se) and specificity (Sp) analysis on the test segments showed values of Se = 3.2%, Sp = 66.0% and Se = 96.8%, Sp = 98.7% in the case of unfiltered and filtered signals during CPR. In conclusion, it was shown that the proposed method can be a useful tool to exactly diagnose the ECG rhythm during the CPR.

An Enhanced Adaptive Filtering Method for Suppressing Cardiopulmonary Resuscitation Artifact.

Cardiopulmonary resuscitation (CPR) must be interrupted for reliable rhythm analysis in current automatic external defibrillators because of artifacts produced by chest compressions. However, interruptions in CPR adversely affect the restoration of spontaneous circulation and survival. Suppressing CPR artifacts by digital signal processing techniques is a promising method to enable rhythm analysis during chest compressions, which would eliminate CPR interruptions for rhythm analysis. Although numerous methods have been developed to suppress CPR artifacts, the accuracy of rhythm analysis is still inadequate due to the residual artifact components in the filtered signal. This study proposes an enhanced adaptive filtering method to suppress CPR artifacts. A total of 183 shockable and 453 nonshockable segments of ECG signal, together with CPR-related reference signal, were extracted from 233 out of hospital cardiac arrest patients. The method was optimized on a training set with 85 shockable and 211 nonshockable segments, and evaluated on a testing set with 98 shockable and 242 nonshockable segments. Compared with artifact corrupted ECG signals, the signal-to-noise ratio (SNR) increased from -9.8±12.5 to 11.2±11.8dB, and the accuracy was improved from 74.1% to 92.0% after filtering with the proposed method. Compared with the traditional adaptive filter, the SNR was improved by 1.7dB and the accuracy was improved by 5.6 points. These results indicated that the proposed method could effectively suppress the chest compression related artifacts and improve the accuracy of rhythm analysis during uninterrupted CPR.

A new method without reference channels used for ventricular fibrillation detection during cardiopulmonary resuscitation.

Ventricular fibrillation (VF) is observed as the initial rhythm in the majority of patients suffering from sudden cardiac arrest. It is vitally important to accurately recognize the initial VF rhythm and then implement electrical defibrillation. However, artifacts produced by chest compression during cardiopulmonary resuscitation (CPR) make the VF detection algorithms utilized by current automated external defibrillators (AEDs) unreliable. CPR must be traditionally interrupted for a reliable diagnosis. However, interruptions in chest compression have a deleterious effect on the success of defibrillation. The elimination of the CPR artifacts would enable compressions to continue during AED VF detection and thereby increase the likelihood of resuscitation success. We have estimated a model of this artifact by adaptively incorporating noise-assisted multivariate empirical mode decomposition (NA-MEMD) and least mean squares (LMS) and then removing the artifact from the corrupted ECGs. The simulation experiment indicated that the CPR artifact could be accurately modeled without any reference channels. We constructed a BP neural network to evaluate the results. A total of 372 VF and 645 normal sinus rhythm (SR) ECG samples were included in the analysis, and 24 CPR artifact signals were used to construct corrupted ECGs. The results indicated that at different SNR levels ranging from 0 to -12dB, the sensitivity and specificity were always above 95 and 80%, respectively.

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.