Automatic identification of return of spontaneous circulation during cardiopulmonary resuscitation

Extracting physiologic and clinical data from defibrillators for research purposes to improve treatment for patients in cardiac arrest

Background

A defibrillator should be connected to all patients receiving cardiopulmonary resuscitation (CPR) to allow early defibrillation. The defibrillator will collect signal data such as the electrocardiogram (ECG), thoracic impedance and end-tidal CO2, which allows for research on how patients demonstrate different responses to CPR. The aim of this review is to give an overview of methodological challenges and opportunities in using defibrillator data for research.

Methods

The successful collection of defibrillator files has several challenges. There is no scientific standard on how to store such data, which have resulted in several proprietary industrial solutions. The data needs to be exported to a software environment where signal filtering and classifications of ECG rhythms can be performed. This may be automated using different algorithms and artificial intelligence (AI). The patient can be classified being in ventricular fibrillation or -tachycardia, asystole, pulseless electrical activity or having obtained return of spontaneous circulation. How this dynamic response is time-dependent and related to covariates can be handled in several ways. These include Aalen's linear model, Weibull regression and joint models.

Conclusions

The vast amount of signal data from defibrillator represents promising opportunities for the use of AI and statistical analysis to assess patient response to CPR. This may provide an epidemiologic basis to improve resuscitation guidelines and give more individualized care. We suggest that an international working party is initiated to facilitate a discussion on how open formats for defibrillator data can be accomplished, that obligates industrial partners to further develop their current technological solutions.

Artificial intelligence in emergency medicine. A systematic literature review

Motivation and objective: Emergency medicine is becoming a popular application area for artificial intelligence methods but remains less investigated than other healthcare branches. The need for time-sensitive decision-making on the basis of high data volumes makes the use of quantitative technologies inevitable. However, the specifics of healthcare regulations impose strict requirements for such applications. Published contributions cover separate parts of emergency medicine and use disparate data and algorithms. This study aims to systematize the relevant contributions, investigate the main obstacles to artificial intelligence applications in emergency medicine, and propose directions for further studies.

METHODS

The contributions selection process was conducted with systematic electronic databases querying and filtering with respect to established exclusion criteria. Among the 380 papers gathered from IEEE Xplore, ACM Digital Library, Springer Library, ScienceDirect, and Nature databases 116 were considered to be a part of the survey. The main features of the selected papers are the focus on emergency medicine and the use of machine learning or deep learning algorithms.

FINDINGS AND DISCUSSION

The selected papers were classified into two branches: diagnostics-specific and triage-specific. The former ones are focused on either diagnosis prediction or decision support. The latter covers such applications as mortality, outcome, admission prediction, condition severity estimation, and urgent care prediction. The observed contributions are highly specialized within a single disease or medical operation and often use privately collected retrospective data, making them incomparable. These and other issues can be addressed by creating an end-to-end solution based on human-machine interaction.

CONCLUSION

Artificial intelligence applications are finding their place in emergency medicine, while most of the corresponding studies remain isolated and lack higher generalization and more sophisticated methodology, which can be a matter of forthcoming improvements.

Artificial intelligence and machine learning in prehospital emergency care: A scoping review

Our scoping review provides a comprehensive analysis of the landscape of artificial intelligence (AI) applications in prehospital emergency care (PEC). It contributes to the field by highlighting the most studied AI applications and identifying the most common methodological approaches across 106 included studies. The findings indicate a promising future for AI in PEC, with many unique use cases, such as prognostication, demand prediction, resource optimization, and the Internet of Things continuous monitoring systems. Comparisons with other approaches showed AI outperforming clinicians and non-AI algorithms in most cases. However, most studies were internally validated and retrospective, highlighting the need for rigorous prospective validation of AI applications before implementation in clinical settings. We identified knowledge and methodological gaps using an evidence map, offering a roadmap for future investigators. We also discussed the significance of explainable AI for establishing trust in AI systems among clinicians and facilitating real-world validation of AI models.

The success rate of cardiopulmonary resuscitation and its correlated factors in patients with emergency prehospital cardiac arrest

To assess the initial success rate and its correlated factors on cardiopulmonary resuscitation (CPR) in emergency prehospital cardiac arrest patients. The clinical information of 429 patients with cardiac arrest who underwent prehospital CPR in the fourth hospital of Hebei Medical University from Jan 2020 to Apr 2022 were evaluated. The patients were divided into the successful group (ROSC, n = 25) and the unsuccessful group (non-ROSC, n = 404) according to whether the autonomous circulation (ROSC) was resumed. The univariate analysis was performed to evaluate the differences in age, the start time of CPR, the application of electric defibrillation, and other related data between the two groups. The multivariate analysis evaluated protective factors affecting CPR's success in prehospital cardiac arrest patients. Patients with cardiogenic causes had the highest success rate of cardiopulmonary resuscitation. The causes of traffic accidents and drowning account for a low proportion. Furthermore, the median CPR length was 25.0 min, alternating from 1.5 to 64 mi. The univariate analysis revealed that age, the start time of CPR, application of electric defibrillation, and adrenaline dosage were correlated with CPR attempts (p < 0.05). Multivariate logistic regression analysis showed that the age of patients with prehospital CA, the location of prehospital CA, etiology, bystander CPR, CPR start time, defibrillation start time, tracheal intubation time, type of rhythm before resuscitation, adrenaline dosage <5 mg, and adrenaline administration time were all the influencing factors of prehospital CPR success (p < 0.01). The factors affecting CPR's success rate in prehospital CA patients are complicated. Establishing a few procedures to diminish the incidence of these risk factors is crucial.

Machine learning and feature engineering for predicting pulse presence during chest compressions

Current resuscitation protocols require pausing chest compressions during cardiopulmonary resuscitation (CPR) to check for a pulse. However, pausing CPR when a patient is pulseless can worsen patient outcomes. Our objective was to design and evaluate an ECG-based algorithm that predicts pulse presence with or without CPR. We evaluated 383 patients being treated for out-of-hospital cardiac arrest with real-time ECG, impedance and audio recordings. Paired ECG segments having an organized rhythm immediately preceding a pulse check (during CPR) and during the pulse check (without CPR) were extracted. Patients were randomly divided into 60% training and 40% test groups. From training data, we developed an algorithm to predict the clinical pulse presence based on the wavelet transform of the bandpass-filtered ECG. Principal component analysis was used to reduce dimensionality, and we then trained a linear discriminant model using three principal component modes as input features. Overall, 38% (351/912) of checks had a spontaneous pulse. AUCs for predicting pulse presence with and without CPR on test data were 0.84 (95% CI (0.80, 0.88)) and 0.89 (95% CI (0.86, 0.92)), respectively. This ECG-based algorithm demonstrates potential to improve resuscitation by predicting the presence of a spontaneous pulse without pausing CPR with moderate accuracy.

Methodology and framework for the analysis of cardiopulmonary resuscitation quality in large and heterogeneous cardiac arrest datasets

BACKGROUND

Out-of-hospital cardiac arrest (OHCA) data debriefing and clinical research often require the retrospective analysis of large datasets containing defibrillator files from different vendors and clinical annotations by the emergency medical services.

AIM

To introduce and evaluate a methodology to automatically extract cardiopulmonary resuscitation (CPR) quality data in a uniform and systematic way from OHCA datasets from multiple heterogeneous sources.

METHODS

A dataset of 2236 OHCA cases from multiple defibrillator models and manufacturers was analyzed. Chest compressions were automatically identified using the thoracic impedance and compression depth signals. Device event time-stamps and clinical annotations were used to set the start and end of the analysis interval, and to identify periods with spontaneous circulation. A manual audit of the automatic annotations was conducted and used as gold standard. Chest compression fraction (CCF), rate (CCR) and interruption ratio were computed as CPR quality variables. The unsigned error between the automated procedure and the gold standard was calculated.

RESULTS

Full-episode median errors below 2% in CCF, 1 min^-1 in CCR, and 1.5% in interruption ratio, were measured for all signals and devices. The proportion of cases with large errors (>10% in CCF and interruption ratio, and >10 min^-1 in CCR) was below 10%. Errors were lower for shorter sub-intervals of interest, like the airway insertion interval.

CONCLUSIONS

An automated methodology was validated to accurately compute CPR metrics in large and heterogeneous OHCA datasets. Automated processing of defibrillator files and the associated clinical annotations enables the aggregation and analysis of CPR data from multiple sources.

A Machine Learning Model for the Prognosis of Pulseless Electrical Activity during Out-of-Hospital Cardiac Arrest

Pulseless electrical activity (PEA) is characterized by the disassociation of the mechanical and electrical activity of the heart and appears as the initial rhythm in 20–30% of out-of-hospital cardiac arrest (OHCA) cases. Predicting whether a patient in PEA will convert to return of spontaneous circulation (ROSC) is important because different therapeutic strategies are needed depending on the type of PEA. The aim of this study was to develop a machine learning model to differentiate PEA with unfavorable (unPEA) and favorable (faPEA) evolution to ROSC. An OHCA dataset of 1921 5s PEA signal segments from defibrillator files was used, 703 faPEA segments from 107 patients with ROSC and 1218 unPEA segments from 153 patients with no ROSC. The solution consisted of a signal-processing stage of the ECG and the thoracic impedance (TI) and the extraction of the TI circulation component (ICC), which is associated with ventricular wall movement. Then, a set of 17 features was obtained from the ECG and ICC signals, and a random forest classifier was used to differentiate faPEA from unPEA. All models were trained and tested using patientwise and stratified 10-fold cross-validation partitions. The best model showed a median (interquartile range) area under the curve (AUC) of 85.7(9.8)% and a balance accuracy of 78.8(9.8)%, improving the previously available solutions at more than four points in the AUC and three points in balanced accuracy. It was demonstrated that the evolution of PEA can be predicted using the ECG and TI signals, opening the possibility of targeted PEA treatment in OHCA.

Multimodal Algorithms for the Classification of Circulation States During Out-of-Hospital Cardiac Arrest

GOAL

Identifying the circulation state during out-of-hospital cardiac arrest (OHCA) is essential to determine what life-saving therapies to apply. Currently algorithms discriminate circulation (pulsed rhythms, PR) from no circulation (pulseless electrical activity, PEA), but PEA can be classified into true (TPEA) and pseudo (PPEA) depending on cardiac contractility. This study introduces multi-class algorithms to automatically determine circulation states during OHCA using the signals available in defibrillators.

METHODS

A cohort of 60 OHCA cases were used to extract a dataset of 2506 5-s segments, labeled as PR (1463), PPEA (364) and TPEA (679) using the invasive blood pressure, experimentally recorded through a radial/femoral cannulation. A multimodal algorithm using features obtained from the electrocardiogram, the thoracic impedance and the capnogram was designed. A random forest model was trained to discriminate three (TPEA/PPEA/PR) and two (PEA/PR) circulation states. The models were evaluated using repeated patient-wise 5-fold cross-validation, with the unweighted mean of sensitivities (UMS) and F ₁-score as performance metrics.

RESULTS

The best model for 3-class had a median (interquartile range, IQR) UMS and F ₁ of 69.0% (68.0-70.1) and 61.7% (61.0-62.5), respectively. The best two class classifier had median (IQR) UMS and F ₁ of 83.9% (82.9-84.5) and 76.2% (75.0-76.9), outperforming all previous proposals in over 3-points in UMS.

CONCLUSIONS

The first multiclass OHCA circulation state classifier was demonstrated. The method improved previous algorithms for binary pulse/no-pulse decisions.

SIGNIFICANCE

Automatic multiclass circulation state classification during OHCA could contribute to improve cardiac arrest therapy and improve survival rates.

Assessment of the evolution of end-tidal carbon dioxide within chest compression pauses to detect restoration of spontaneous circulation

Background

Measurement of end-tidal CO₂ (ETCO₂) can help to monitor circulation during cardiopulmonary resuscitation (CPR). However, early detection of restoration of spontaneous circulation (ROSC) during CPR using waveform capnography remains a challenge. The aim of the study was to investigate if the assessment of ETCO₂ variation during chest compression pauses could allow for ROSC detection. We hypothesized that a decay in ETCO₂ during a compression pause indicates no ROSC while a constant or increasing ETCO2 indicates ROSC.

Methods

We conducted a retrospective analysis of adult out-of-hospital cardiac arrest (OHCA) episodes treated by the advanced life support (ALS). Continuous chest compressions and ventilations were provided manually. Segments of capnography signal during pauses in chest compressions were selected, including at least three ventilations and with durations less than 20 s. Segments were classified as ROSC or non-ROSC according to case chart annotation and examination of the ECG and transthoracic impedance signals. The percentage variation of ETCO₂ between consecutive ventilations was computed and its average value, ΔET_avg, was used as a single feature to discriminate between ROSC and non-ROSC segments.

Results

A total of 384 segments (130 ROSC, 254 non-ROSC) from 205 OHCA patients (30.7% female, median age 66) were analyzed. Median (IQR) duration was 16.3 (12.9,18.1) s. ΔET_avg was 0.0 (-0.7, 0.9)% for ROSC segments and -11.0 (-14.1, -8.0)% for non-ROSC segments (p < 0.0001). Best performance for ROSC detection yielded a sensitivity of 95.4% (95% CI: 90.1%, 98.1%) and a specificity of 94.9% (91.4%, 97.1%) for all ventilations in the segment. For the first 2 ventilations, duration was 7.7 (6.0, 10.2) s, and sensitivity and specificity were 90.0% (83.5%, 94.2%) and 89.4 (84.9%, 92.6%), respectively. Our method allowed for ROSC detection during the first compression pause in 95.4% of the patients.

Conclusion

Average percent variation of ETCO₂ during pauses in chest compressions allowed for ROSC discrimination. This metric could help confirm ROSC during compression pauses in ALS settings.

ECG-based Random Forest Classifier for Cardiac Arrest Rhythms

Rhythm annotation of out-of-hospital cardiac episodes (OHCA) is key for a better understanding of the interplay between resuscitation therapy and OHCA patient outcome. OHCA rhythms are classified in five categories, asystole (AS), pulseless electrical activity (PEA), pulsed rhythms (PR), ventricular fibrillation (VF) and ventricular tachycardia (VT). Manual OHCA annotation by expert clinicians is onerous and time consuming, so there is a need for accurate and automatic OHCA rhythm annotation methods. For this study 852 OHCA episodes of patients treated with Automated External Defibrillators (AED) by the Emergency Medical Services of the Basque Country were analyzed. Six expert clinicians reviewed the electrocardiogram (ECG) of 4214 AED rhythm analyses and annotated the rhythm. Their consensus decision was used as ground truth. There were a total of 2418 AS, 294 PR, 1008 PEA, 472 VF and 22 VT. The ECG analysis intervals were extracted and used to develop an automatic rhythm annotator. Data was partitioned patient-wise into training (70%) and test (30%). Performance was evaluated in terms of per class sensitivity (Se) and F-score (F1). The unweighted mean of sensitivity (UMS) and F-score were used as global performance metrics. The classification method is composed of a feature extraction and denoising stage based on the stationary wavelet transform of the ECG, and on a random forest classifier. The best model presented a per rhythm Se/F1 of 95.8/95.7, 43.3/52.2, 85.3/81.3, 94.2/96.1, 81.9/72.2 for AS, PR, PEA, VF and VT, respectively. The UMS for the test set was 80.2%, 2-points above that of previous solutions. This method could be used to retrospectively annotate large OHCA datasets and ameliorate the workload of manual OHCA rhythm annotation.

Convolutional Recurrent Neural Networks to Characterize the Circulation Component in the Thoracic Impedance during Out-of-Hospital Cardiac Arrest

Pulse detection during out-of-hospital cardiac arrest remains challenging for both novel and expert rescuers because current methods are inaccurate and time-consuming. There is still a need to develop automatic methods for pulse detection, where the most challenging scenario is the discrimination between pulsed rhythms (PR, pulse) and pulseless electrical activity (PEA, no pulse). Thoracic impedance (TI) acquired through defibrillation pads has been proven useful for detecting pulse as it shows small fluctuations with every heart beat. In this study we analyse the use of deep learning techniques to detect pulse using only the TI signal. The proposed neural network, composed by convolutional and recurrent layers, outperformed state of the art methods, and achieved a balanced accuracy of 90% for segments as short as 3 s.

On detection of spontaneous pulse by photoplethysmography in cardiopulmonary resuscitation

OBJECTIVE

This work investigates the potential of photoplethysmography (PPG) to detect a spontaneous pulse from the finger, nose or ear in order to support pulse checks during cardiopulmonary resuscitation (CPR).

METHODS

In a prospective single-center cross-sectional study, PPG signals were acquired from cardiac arrest victims who underwent CPR. The PPG signals were analyzed and compared to arterial blood pressure (ABP) signals as a reference during three distranaisco; Date: 2/2/2020; Time:18:44:23inct phases of CPR: compression pauses, on-going compressions and at very low arterial blood pressure. Data analysis was based on a qualitative subjective visual description of similarities of the frequency content of PPG and ABP waveform.

RESULTS

In 9 patients PPG waveforms corresponded to ABP waveforms during normal blood pressures. During ABP in the clinically challenging range of 60 to 90 mmHg and during chest compressions and pauses, PPG continued to resemble ABP, as both signals showed similar frequency components as a result of chest compressions as well as cardiac activity. Altogether 1199 s of PPG data in compression pauses were expected to show a spontaneous pulse, of which 732 s (61%) of data were artifact-free and showed the spontaneous pulse as visible in the ABP.

CONCLUSIONS

PPG signals at all investigated sites can indicate pulse presence at the moment the heart resumes beating as verified via the ABP signal. Therefore, PPG may provide decision support during CPR, especially related to preventing and shortening interruptions for unnecessary pulse checks. This could have impact on CPR outcome and should further be investigated.

Capnography: A support tool for the detection of return of spontaneous circulation in out-of-hospital cardiac arrest

BACKGROUND

Automated detection of return of spontaneous circulation (ROSC) is still an unsolved problem during cardiac arrest. Current guidelines recommend the use of capnography, but most automatic methods are based on the analysis of the ECG and thoracic impedance (TI) signals. This study analysed the added value of EtCO₂ for discriminating pulsed (PR) and pulseless (PEA) rhythms and its potential to detect ROSC.

MATERIALS AND METHODS

A total of 426 out-of-hospital cardiac arrest cases, 117 with ROSC and 309 without ROSC, were analysed. First, EtCO₂ values were compared for ROSC and no ROSC cases. Second, 5098 artefact free 3-s long segments were automatically extracted and labelled as PR (3639) or PEA (1459) using the instant of ROSC annotated by the clinician on scene as gold standard. Machine learning classifiers were designed using features obtained from the ECG, TI and the EtCO₂ value. Third, the cases were retrospectively analysed using the classifier to discriminate cases with and without ROSC.

RESULTS

EtCO₂ values increased significantly from 41 mmHg 3-min before ROSC to 57 mmHg 1-min after ROSC, and EtCO₂ was significantly larger for PR than for PEA, 46 mmHg/20 mmHg (p < 0.05). Adding EtCO₂ to the machine learning models increased their area under the curve (AUC) by over 2 percentage points. The combination of ECG, TI and EtCO₂ had an AUC for the detection of pulse of 0.92. Finally, the retrospective analysis showed a sensitivity and specificity of 96.6% and 94.5% for the detection of ROSC and no-ROSC cases, respectively.

CONCLUSION

Adding EtCO₂ improves the performance of automatic algorithms for pulse detection based on ECG and TI. These algorithms can be used to identify pulse on site, and to retrospectively identify cases with ROSC.

Deep Neural Networks for ECG-Based Pulse Detection during Out-of-Hospital Cardiac Arrest

The automatic detection of pulse during out-of-hospital cardiac arrest (OHCA) is necessary for the early recognition of the arrest and the detection of return of spontaneous circulation (end of the arrest). The only signal available in every single defibrillator and valid for the detection of pulse is the electrocardiogram (ECG). In this study we propose two deep neural network (DNN) architectures to detect pulse using short ECG segments (5 s), i.e., to classify the rhythm into pulseless electrical activity (PEA) or pulse-generating rhythm (PR). A total of 3914 5-s ECG segments, 2372 PR and 1542 PEA, were extracted from 279 OHCA episodes. Data were partitioned patient-wise into training (80%) and test (20%) sets. The first DNN architecture was a fully convolutional neural network, and the second architecture added a recurrent layer to learn temporal dependencies. Both DNN architectures were tuned using Bayesian optimization, and the results for the test set were compared to state-of-the art PR/PEA discrimination algorithms based on machine learning and hand crafted features. The PR/PEA classifiers were evaluated in terms of sensitivity (Se) for PR, specificity (Sp) for PEA, and the balanced accuracy (BAC), the average of Se and Sp. The Se/Sp/BAC of the DNN architectures were 94.1%/92.9%/93.5% for the first one, and 95.5%/91.6%/93.5% for the second one. Both architectures improved the performance of state of the art methods by more than 1.5 points in BAC.

ECG-based pulse detection during cardiac arrest using random forest classifier

Sudden cardiac arrest is one of the leading causes of death in the industrialized world. Pulse detection is essential for the recognition of the arrest and the recognition of return of spontaneous circulation during therapy, and it is therefore crucial for the survival of the patient. This paper introduces the first method based exclusively on the ECG for the automatic detection of pulse during cardiopulmonary resuscitation. Random forest classifier is used to efficiently combine up to nine features from the time, frequency, slope, and regularity analysis of the ECG. Data from 191 cardiac arrest patients was used, and 1177 ECG segments were processed, 796 with pulse and 381 without pulse. A leave-one-patient out cross validation approach was used to train and test the algorithm. The statistical distributions of sensitivity (SE) and specificity (SP) for pulse detection were estimated using 500 patient-wise bootstrap partitions. The mean (std) SE/SP for nine-feature classifier was 88.4 (1.8) %/89.7 (1.4) %, respectively. The designed algorithm only requires 4-s-long ECG segments and could be integrated in any commercial automated external defibrillator. The method permits to detect the presence of pulse accurately, minimizing interruptions in cardiopulmonary resuscitation therapy, and could contribute to improve survival from cardiac arrest.

Circulation assessment by automated external defibrillators during cardiopulmonary resuscitation

AIM

To design and evaluate a simple algorithm able to discriminate pulsatile rhythms from pulseless electrical activity during automated external defibrillator (AED) analysis intervals, using the ECG and the transthoracic impedance (TI) acquired from defibrillation pads.

METHODS

ECG and TI signals from out-of-hospital AED recordings were retrospectively analysed. Experts annotated the cardiac rhythm during AED analysis intervals and at the end of each episode. We developed an algorithm to classify 3-s segments of non-shockable and non-asystole rhythms as either pulsatile rhythm or pulseless electrical activity. The algorithm consisted on a decision tree based on two features: the mean power of the TI segment and the mean cross-power between ECG and TI segments.

RESULTS

From the 302 annotated episodes, 167 contained segments eligible for the study. The circulation detector algorithm presented a sensitivity (ability of detecting pulsatile rhythms) of 98.3% (95% CI: 95.1-100) and a specificity (ability to detect pulseless electrical activity) of 98.4% (95% CI: 97.1-99.8) in the validation subset. Absence of pulsatile rhythm was confirmed during the first AED analysis interval in 98.9% of the episodes, and presence of a pulse was confirmed in the first 3 s of all intervals with annotated return of spontaneous circulation.

CONCLUSION

Accurate automated detection of circulation based on TI and ECG is possible during AED analysis intervals. This functionality could potentially contribute to enhance patient's care by laypersons using AEDs.

An automatic system for the comprehensive retrospective analysis of cardiac rhythms in resuscitation episodes

AIM

An automatic resuscitation rhythm annotator (ARA) would facilitate and enhance retrospective analysis of resuscitation data, contributing to a better understanding of the interplay between therapy and patient response. The objective of this study was to define, implement, and demonstrate an ARA architecture for complete resuscitation episodes, including chest compression pauses (CC-pauses) and chest compression intervals (CC-intervals).

METHODS

We analyzed 126.5h of ECG and accelerometer-based chest-compression depth data from 281 out-of-hospital cardiac arrest (OHCA) patients. Data were annotated by expert reviewers into asystole (AS), pulseless electrical activity (PEA), pulse-generating rhythm (PR), ventricular fibrillation (VF), and ventricular tachycardia (VT). Clinical pulse annotations were based on patient-charts and impedance measurements. An ARA was developed for CC-pauses, and was used in combination with a chest compression artefact removal filter during CC-intervals. The performance of the ARA was assessed in terms of the unweighted mean of sensitivities (UMS).

RESULTS

The UMS of the ARA were 75.0% during CC-pauses and 52.5% during CC-intervals, 55-points and 32.5-points over a random guess (20% for five categories). Filtering increased the UMS during CC-intervals by 5.2-points. Sensitivities for AS, PEA, PR, VF, and VT were 66.8%, 55.8%, 86.5%, 82.1% and 83.8% during CC-pauses; and 51.1%, 34.1%, 58.7%, 86.4%, and 32.1% during CC-intervals.

CONCLUSIONS

A general ARA architecture was defined and demonstrated on a comprehensive OHCA dataset. Results showed that semi-automatic resuscitation rhythm annotation, which may involve further revision/correction by clinicians for quality assurance, is feasible. The performance (UMS) dropped significantly during CC-intervals and sensitivity was lowest for PEA.

ECG-Based Classification of Resuscitation Cardiac Rhythms for Retrospective Data Analysis

OBJECTIVE

There is a need to monitor the heart rhythm in resuscitation to improve treatment quality. Resuscitation rhythms are categorized into: ventricular tachycardia (VT), ventricular fibrillation (VF), pulseless electrical activity (PEA), asystole (AS), and pulse-generating rhythm (PR). Manual annotation of rhythms is time-consuming and infeasible for large datasets. Our objective was to develop ECG-based algorithms for the retrospective and automatic classification of resuscitation cardiac rhythms.

METHODS

The dataset consisted of 1631 3-s ECG segments with clinical rhythm annotations, obtained from 298 out-of-hospital cardiac arrest patients. In total, 47 wavelet- and time-domain-based features were computed from the ECG. Features were selected using a wrapper-based feature selection architecture. Classifiers based on Bayesian decision theory, k-nearest neighbor, k-local hyperplane distance nearest neighbor, artificial neural network (ANN), and ensemble of decision trees were studied.

RESULTS

The best results were obtained for ANN classifier with Bayesian regularization backpropagation training algorithm with 14 features, which forms the proposed algorithm. The overall accuracy for the proposed algorithm was 78.5%. The sensitivities (and positive-predictive-values) for AS, PEA, PR, VF, and VT were 88.7% (91.0%), 68.9% (70.4%), 65.9% (69.0%), 86.2% (83.8%), and 78.8% (72.9%), respectively.

CONCLUSIONS

The results demonstrate that it is possible to classify resuscitation cardiac rhythms automatically, but the accuracy for the organized rhythms (PEA and PR) is low.

SIGNIFICANCE

We have made an important step toward making classification of resuscitation rhythms more efficient in the sense of minimal feedback from human experts.

Circulation detection using the electrocardiogram and the thoracic impedance acquired by defibrillation pads

AIM

To develop and evaluate a method to detect circulation in the presence of organized rhythms (ORs) during resuscitation using signals acquired by defibrillation pads.

METHODS

Segments containing electrocardiogram (ECG) and thoracic impedance (TI) signals free of artifacts were used. The ECG corresponded to ORs classified as pulseless electrical activity (PEA) or pulse-generating rhythm (PR). A first dataset containing 1091 segments was split into training and test sets to develop and validate the circulation detector. The method processed ECG and TI to obtain the impedance circulation component (ICC). Morphological features were extracted from ECG and ICC, and combined into a classifier to discriminate between PEA and PR. The performance of the method was evaluated in terms of sensitivity (PR) and specificity (PEA). A second dataset (86 segments from different patients) was used to assess two application of the method: confirmation of arrest by recognizing absence of circulation during ORs and detection of return of spontaneous circulation (ROSC) during resuscitation. In both cases, time to confirmation of arrest/ROSC was determined.

RESULTS

The method showed a sensitivity/specificity of 92.1%/90.3% and 92.2%/91.9% for training and test sets respectively. The method confirmed cardiac arrest with a specificity of 93.3% with a median delay of 0s after the first OR annotation. ROSC was detected with a sensitivity of 94.4% with a median delay of 57s from ROSC onset.

CONCLUSION

The method showed good performance, and can be reliably used to distinguish perfusing from non-perfusing ORs.

Photoplethysmography-based algorithm for detection of cardiogenic output during cardiopulmonary resuscitation

Detecting return of spontaneous circulation (ROSC) during cardiopulmonary resuscitation (CPR) is challenging, time consuming, and requires interrupting chest compressions. Based on automated-CPR porcine data, we have developed an algorithm to support ROSC detection, which detects cardiogenic output during chest compressions via a photoplethysmography (PPG) signal. The algorithm can detect palpable and impalpable spontaneous pulses. A compression-free PPG signal which estimates the spontaneous pulse waveform, was obtained by subtracting the compression component, modeled by a harmonic series. The fundamental frequency of this series was the compression rate derived from the transthoracic impedance signal measured between the defibrillation pads. The amplitudes of the harmonic components were obtained via a least mean-square algorithm. The frequency spectrum of the compression-free PPG signal was estimated via an autoregressive model, and the relationship between the spectral peaks was analyzed to identify the pulse rate (PR). Resumed cardiogenic output could also be detected from a decrease in the baseline of the PPG signal, presumably caused by a redistribution of blood volume to the periphery. The algorithm indicated cardiogenic output when a PR or a redistribution of blood volume was detected. The algorithm indicated cardiogenic output with 94% specificity and 69% sensitivity compared to the retrospective ROSC detection of nine clinicians. Results showed that ROSC detection can be supported by combining the compression-free PPG signal with an indicator based on the detected PR and redistribution of blood volume.

Towards the automated analysis and database development of defibrillator data from cardiac arrest

BACKGROUND

During resuscitation of cardiac arrest victims a variety of information in electronic format is recorded as part of the documentation of the patient care contact and in order to be provided for case review for quality improvement. Such review requires considerable effort and resources. There is also the problem of interobserver effects.

OBJECTIVE

We show that it is possible to efficiently analyze resuscitation episodes automatically using a minimal set of the available information.

METHODS AND RESULTS

A minimal set of variables is defined which describe therapeutic events (compression sequences and defibrillations) and corresponding patient response events (annotated rhythm transitions). From this a state sequence representation of the resuscitation episode is constructed and an algorithm is developed for reasoning with this representation and extract review variables automatically. As a case study, the method is applied to the data abstraction process used in the King County EMS. The automatically generated variables are compared to the original ones with accuracies ≥ 90% for 18 variables and ≥ 85% for the remaining four variables.

CONCLUSIONS

It is possible to use the information present in the CPR process data recorded by the AED along with rhythm and chest compression annotations to automate the episode review.

Detection of a spontaneous pulse in photoplethysmograms during automated cardiopulmonary resuscitation in a porcine model

INTRODUCTION

Reliable, non-invasive detection of return of spontaneous circulation (ROSC) with minimal interruptions to chest compressions would be valuable for high-quality cardiopulmonary resuscitation (CPR). We investigated the potential of photoplethysmography (PPG) to detect the presence of a spontaneous pulse during automated CPR in an animal study.

METHODS

Twelve anesthetized pigs were instrumented to monitor circulatory and respiratory parameters. Here we present the simultaneously recorded PPG and arterial blood pressure (ABP) signals. Ventricular fibrillation was induced, followed by 20 min of automated CPR and subsequent defibrillation. After defibrillation, pediatric-guidelines-style life support was given in cycles of 2 min. PPG and ABP waveforms were recorded during all stages of the protocol. Raw PPG waveforms were acquired with a custom-built photoplethysmograph controlling a commercial reflectance pulse oximetry probe attached to the nose. ABP was measured in the aorta.

RESULTS

In nine animals ROSC was achieved. Throughout the protocol, PPG and ABP frequency content showed strong resemblance. We demonstrate that (1) the PPG waveform allows for the detection of a spontaneous pulse during ventilation pauses, and that (2) frequency analysis of the PPG waveform allows for the detection of a spontaneous pulse and the determination of the pulse rate, even during ongoing chest compressions, if the pulse and compression rates are sufficiently distinct.

CONCLUSIONS

These results demonstrate the potential of PPG as a non-invasive means to detect pulse presence or absence, as well as pulse rate during CPR.

Dynamic effects of adrenaline (epinephrine) in out-of-hospital cardiac arrest with initial pulseless electrical activity (PEA)

BACKGROUND

In cardiac arrest, pulseless electrical activity (PEA) is a challenging clinical syndrome. In a randomized study comparing intravenous (i.v.) access and drugs versus no i.v. access or drugs during advanced life support (ALS), adrenaline (epinephrine) improved return of spontaneous circulation (ROSC) in patients with PEA. Originating from this study, we investigated the time-dependent effects of adrenaline on clinical state transitions in patients with initial PEA, using a non-parametric multi-state statistical model.

METHODS AND RESULTS

Patients with available defibrillator recordings were included, of whom 101 received adrenaline and 73 did not. There were significantly more state transitions in the adrenaline group than in the no-adrenaline group (rate ratio = 1.6, p<0.001). Adrenaline markedly increased the rate of transition from PEA to ROSC during ALS and slowed the rate of being declared dead; e.g. by 20 min 20% of patients in the adrenaline group had been declared dead and 25% had obtained ROSC, whereas 50% in the no-adrenaline group have been declared dead and 15% had obtained ROSC. The differential effect of adrenaline could be seen after approx. 10 min of ALS for most transitions. For both groups the probability of deteriorating from PEA to asystole was highest during the first 15 min. Adrenaline increased the rate of transition from PEA to ventricular fibrillation or -tachycardia (VF/VT), and from ROSC to VF/VT.

CONCLUSIONS

Adrenaline has notable clinical effects during ALS in patients with initial PEA. The drug extends the time window for ROSC to develop, but also renders the patient more unstable. Further research should investigate the optimal dose, timing and mode of adrenaline administration during ALS.

Detection of cardiac arrest using a simplified frequency analysis of the impedance cardiogram recorded from defibrillator pads

An algorithm based only on the impedance cardiogram (ICG) recorded through two defibrillation pads, using the strongest frequency component and amplitude, incorporated into a defibrillator could determine circulatory arrest and reduce delays in starting cardiopulmonary resuscitation (CPR). Frequency analysis of the ICG signal is carried out by integer filters on a sample by sample basis. They are simpler, lighter and more versatile when compared to the FFT. This alternative approach, although less accurate, is preferred due to the limited processing capacity of devices that could compromise real time usability of the FFT. These two techniques were compared across a data set comprising 13 cases of cardiac arrest and 6 normal controls. The best filters were refined on this training set and an algorithm for the detection of cardiac arrest was trained on a wider data set. The algorithm was finally tested on a validation set. The ICG was recorded in 132 cardiac arrest patients (53 training, 79 validation) and 97 controls (47 training, 50 validation): the diagnostic algorithm indicated cardiac arrest with a sensitivity of 81.1% (77.6-84.3) and specificity of 97.1% (96.7-97.4) for the validation set (95% confidence intervals). Automated defibrillators with integrated ICG analysis have the potential to improve emergency care by lay persons enabling more rapid and appropriate initiation of CPR and when combined with ECG analysis they could improve on the detection of cardiac arrest.

Using ontologies to integrate and share resuscitation data from diverse medical devices

OBJECTIVE

To propose a method for standardised data representation and demonstrate a technology that makes it possible to translate data from device dependent formats to this standard representation format.

METHODS AND RESULTS

Outcome statistics vary between emergency medical systems organising resuscitation services. Such differences indicate a potential for improvement by identifying factors affecting outcome, but data subject to analysis have to be comparable. Modern technology for communicating information makes it possible to structure, store and transfer data flexibly. Ontologies describe entities in the world and how they relate. Letting different computer systems refer to the same ontology results in a common understanding on data content. Information on therapy such as shock delivery, chest compressions and ventilation should be defined and described in a standardised ontology to enable comparison and combining data from diverse sources. By adding rules and logic data can be merged and combined in new ways to produce new information. An example ontology is designed to demonstrate the feasibility and value of such a standardised structure.

CONCLUSIONS

The proposed technology makes possible capturing and storing of data from different devices in a structured and standardised format. Data can easily be transformed to this standardised format, compared and combined independent of the original structure.