Prediction of cardiac arrest in critically ill patients presenting to the emergency department using a machine learning score incorporating heart rate variability compared with the modified early warning score

Artificial Intelligence in Pediatric Emergency Medicine: Applications, Challenges, and Future Perspectives

The dawn of Artificial intelligence (AI) in healthcare stands as a milestone in medical innovation. Different medical fields are heavily involved, and pediatric emergency medicine is no exception. We conducted a narrative review structured in two parts. The first part explores the theoretical principles of AI, providing all the necessary background to feel confident with these new state-of-the-art tools. The second part presents an informative analysis of AI models in pediatric emergencies. We examined PubMed and Cochrane Library from inception up to April 2024. Key applications include triage optimization, predictive models for traumatic brain injury assessment, and computerized sepsis prediction systems. In each of these domains, AI models outperformed standard methods. The main barriers to a widespread adoption include technological challenges, but also ethical issues, age-related differences in data interpretation, and the paucity of comprehensive datasets in the pediatric context. Future feasible research directions should address the validation of models through prospective datasets with more numerous sample sizes of patients. Furthermore, our analysis shows that it is essential to tailor AI algorithms to specific medical needs. This requires a close partnership between clinicians and developers. Building a shared knowledge platform is therefore a key step.

Wireless Battery-free and Fully Implantable Organ Interfaces

Advances in soft materials, miniaturized electronics, sensors, stimulators, radios, and battery-free power supplies are resulting in a new generation of fully implantable organ interfaces that leverage volumetric reduction and soft mechanics by eliminating electrochemical power storage. This device class offers the ability to provide high-fidelity readouts of physiological processes, enables stimulation, and allows control over organs to realize new therapeutic and diagnostic paradigms. Driven by seamless integration with connected infrastructure, these devices enable personalized digital medicine. Key to advances are carefully designed material, electrophysical, electrochemical, and electromagnetic systems that form implantables with mechanical properties closely matched to the target organ to deliver functionality that supports high-fidelity sensors and stimulators. The elimination of electrochemical power supplies enables control over device operation, anywhere from acute, to lifetimes matching the target subject with physical dimensions that supports imperceptible operation. This review provides a comprehensive overview of the basic building blocks of battery-free organ interfaces and related topics such as implantation, delivery, sterilization, and user acceptance. State of the art examples categorized by organ system and an outlook of interconnection and advanced strategies for computation leveraging the consistent power influx to elevate functionality of this device class over current battery-powered strategies is highlighted.

Real-time machine learning model to predict in-hospital cardiac arrest using heart rate variability in ICU

Predicting in-hospital cardiac arrest in patients admitted to an intensive care unit (ICU) allows prompt interventions to improve patient outcomes. We developed and validated a machine learning-based real-time model for in-hospital cardiac arrest predictions using electrocardiogram (ECG)-based heart rate variability (HRV) measures. The HRV measures, including time/frequency domains and nonlinear measures, were calculated from 5 min epochs of ECG signals from ICU patients. A light gradient boosting machine (LGBM) algorithm was used to develop the proposed model for predicting in-hospital cardiac arrest within 0.5-24 h. The LGBM model using 33 HRV measures achieved an area under the receiver operating characteristic curve of 0.881 (95% CI: 0.875-0.887) and an area under the precision-recall curve of 0.104 (95% CI: 0.093-0.116). The most important feature was the baseline width of the triangular interpolation of the RR interval histogram. As our model uses only ECG data, it can be easily applied in clinical practice.

Machine Learning Models Versus the National Early Warning Score System for Predicting Deterioration: Retrospective Cohort Study in the United Arab Emirates

BACKGROUND

Early warning score systems are widely used for identifying patients who are at the highest risk of deterioration to assist clinical decision-making. This could facilitate early intervention and consequently improve patient outcomes; for example, the National Early Warning Score (NEWS) system, which is recommended by the Royal College of Physicians in the United Kingdom, uses predefined alerting thresholds to assign scores to patients based on their vital signs. However, there is limited evidence of the reliability of such scores across patient cohorts in the United Arab Emirates.

OBJECTIVE

Our aim in this study was to propose a data-driven model that accurately predicts in-hospital deterioration in an inpatient cohort in the United Arab Emirates.

METHODS

We conducted a retrospective cohort study using a real-world data set that consisted of 16,901 unique patients associated with 26,073 inpatient emergency encounters and 951,591 observation sets collected between April 2015 and August 2021 at a large multispecialty hospital in Abu Dhabi, United Arab Emirates. The observation sets included routine measurements of heart rate, respiratory rate, systolic blood pressure, level of consciousness, temperature, and oxygen saturation, as well as whether the patient was receiving supplementary oxygen. We divided the data set of 16,901 unique patients into training, validation, and test sets consisting of 11,830 (70%; 18,319/26,073, 70.26% emergency encounters), 3397 (20.1%; 5206/26,073, 19.97% emergency encounters), and 1674 (9.9%; 2548/26,073, 9.77% emergency encounters) patients, respectively. We defined an adverse event as the occurrence of admission to the intensive care unit, mortality, or both if the patient was admitted to the intensive care unit first. On the basis of 7 routine vital signs measurements, we assessed the performance of the NEWS system in detecting deterioration within 24 hours using the area under the receiver operating characteristic curve (AUROC). We also developed and evaluated several machine learning models, including logistic regression, a gradient-boosting model, and a feed-forward neural network.

RESULTS

In a holdout test set of 2548 encounters with 95,755 observation sets, the NEWS system achieved an overall AUROC value of 0.682 (95% CI 0.673-0.690). In comparison, the best-performing machine learning models, which were the gradient-boosting model and the neural network, achieved AUROC values of 0.778 (95% CI 0.770-0.785) and 0.756 (95% CI 0.749-0.764), respectively. Our interpretability results highlight the importance of temperature and respiratory rate in predicting patient deterioration.

CONCLUSIONS

Although traditional early warning score systems are the dominant form of deterioration prediction models in clinical practice today, we strongly recommend the development and use of cohort-specific machine learning models as an alternative. This is especially important in external patient cohorts that were unseen during model development.

Artificial Intelligence in Ventricular Arrhythmias and Sudden Death

Sudden cardiac arrest due to lethal ventricular arrhythmias is a major cause of mortality worldwide and results in more years of potential life lost than any individual cancer. Most of these sudden cardiac arrest events occur unexpectedly in individuals who have not been identified as high-risk due to the inadequacy of current risk stratification tools. Artificial intelligence tools are increasingly being used to solve complex problems and are poised to help with this major unmet need in the field of clinical electrophysiology. By leveraging large and detailed datasets, artificial intelligence-based prediction models have the potential to enhance the risk stratification of lethal ventricular arrhythmias. This review presents a synthesis of the published literature and a discussion of future directions in this field.

Artificial Intelligence in Resuscitation: A Scoping Review

Introduction: Cardiac arrest is a significant cause of premature mortality and severe disability. Despite the death rate steadily decreasing over the previous decade, only 22% of survivors achieve good clinical status and only 25% of patients survive until their discharge from the hospital. The objective of this scoping review was to review relevant AI modalities and the main potential applications of AI in resuscitation. Methods: We conducted the literature search for related studies in PubMed, EMBASE, and Google Scholar. We included peer-reviewed publications and articles in the press, pooling and characterizing the data by their model types, goals, and benefits. Results: After identifying 268 original studies, we chose 59 original studies (reporting 1,817,419 patients) to include in the qualitative synthesis. AI-based methods appear to be superior to traditional methods in achieving high-level performance. Conclusion: AI might be useful in predicting cardiac arrest, heart rhythm disorders, and post-cardiac arrest outcomes, as well as in the delivery of drone-delivered defibrillators and notification of dispatchers. AI-powered technologies could be valuable assistants to continuously track patient conditions. Healthcare professionals should assist in the research and development of AI-powered technologies as well as their implementation into clinical practice.

Machine learning to predict in-hospital cardiac arrest from patients presenting to the emergency department

In-hospital cardiac arrest (IHCA) in the emergency department (ED) is not uncommon but often fatal. Using the machine learning (ML) approach, we sought to predict ED-based IHCA (EDCA) in patients presenting to the ED based on triage data. We retrieved 733,398 ED records from a tertiary teaching hospital over a 7 year period (Jan. 1, 2009-Dec. 31, 2015). We included only adult patients (≥ 18 y) and excluded cases presenting as out-of-hospital cardiac arrest. Primary outcome (EDCA) was identified via a resuscitation code. Patient demographics, triage data, and structured chief complaints (CCs), were extracted. Stratified split was used to divide the dataset into the training and testing cohort at a 3-to-1 ratio. Three supervised ML models were trained and performances were evaluated and compared to the National Early Warning Score 2 (NEWS2) and logistic regression (LR) model by the area under the receiver operating characteristic curve (AUC). We included 316,465 adult ED records for analysis. Of them, 636 (0.2%) developed EDCA. Of the constructed ML models, Random Forest outperformed the others with the best AUC result (0.931, 95% CI 0.911-0.949), followed by Gradient Boosting (0.930, 95% CI 0.909-0.948) and Extra Trees classifier (0.915, 95% CI 0.892-0.936). Although the differences between each of ML models and LR (AUC: 0.905, 95% CI 0.882-0.926) were not significant, all constructed ML models performed significantly better than using the NEWS2 scoring system (AUC 0.678, 95% CI 0.635-0.722). Our ML models showed excellent discriminatory performance to identify EDCA based only on the triage information. This ML approach has the potential to reduce unexpected resuscitation events if successfully implemented in the ED information system.

Machine learning of electrophysiological signals for the prediction of ventricular arrhythmias: systematic review and examination of heterogeneity between studies

BACKGROUND

Ventricular arrhythmia (VA) precipitating sudden cardiac arrest (SCD) is among the most frequent causes of death and pose a high burden on public health systems worldwide. The increasing availability of electrophysiological signals collected through conventional methods (e.g. electrocardiography (ECG)) and digital health technologies (e.g. wearable devices) in combination with novel predictive analytics using machine learning (ML) and deep learning (DL) hold potential for personalised predictions of arrhythmic events.

METHODS

This systematic review and exploratory meta-analysis assesses the state-of-the-art of ML/DL models of electrophysiological signals for personalised prediction of malignant VA or SCD, and studies potential causes of bias (PROSPERO, reference: CRD42021283464). Five electronic databases were searched to identify eligible studies. Pooled estimates of the diagnostic odds ratio (DOR) and summary area under the curve (AUROC) were calculated. Meta-analyses were performed separately for studies using publicly available, ad-hoc datasets, versus targeted clinical data acquisition. Studies were scored on risk of bias by the PROBAST tool.

FINDINGS

2194 studies were identified of which 46 were included in the systematic review and 32 in the meta-analysis. Pooling of individual models demonstrated a summary AUROC of 0.856 (95% CI 0.755-0.909) for short-term (time-to-event up to 72 h) prediction and AUROC of 0.876 (95% CI 0.642-0.980) for long-term prediction (time-to-event up to years). While models developed on ad-hoc sets had higher pooled performance (AUROC 0.919, 95% CI 0.867-0.952), they had a high risk of bias related to the re-use and overlap of small ad-hoc datasets, choices of ML tool and a lack of external model validation.

INTERPRETATION

ML and DL models appear to accurately predict malignant VA and SCD. However, wide heterogeneity between studies, in part due to small ad-hoc datasets and choice of ML model, may reduce the ability to generalise and should be addressed in future studies.

FUNDING

This publication is part of the project DEEP RISK ICD (with project number 452019308) of the research programme Rubicon which is (partly) financed by the Dutch Research Council (NWO). This research is partly funded by the Amsterdam Cardiovascular Sciences (personal grant F.V.Y.T).

Heart Rate Variability Can Detect Blunt Traumatic Brain Injury Within the First Hour

INTRODUCTION

In patients with multi-organ system trauma, the diagnosis of coinciding traumatic brain injury can be difficult due to injuries from the hemorrhagic shock that confound clinical and radiographic signs of traumatic brain injury. In this study, a novel technique using heart rate variability was developed in a porcine model to detect traumatic brain injury early in the setting of hemorrhagic shock without the need for radiographic imaging or clinical exam.

METHODS

A porcine model of hemorrhagic shock was used with an arm of swine receiving hemorrhagic shock alone and hemorrhagic shock with traumatic brain injury. High-resolution heart rate frequencies were collected at different time intervals using waveforms based on voltage delivered from the heart rate monitor. Waveforms were analyzed to assess statistically significant differences between heart rate variability parameters in those with hemorrhagic shock and traumatic brain injury versus those with only hemorrhagic shock. Stochastic analysis was used to assess the validity of results and create a model by machine learning to better assess the presence of traumatic brain injury.

RESULTS

Significant differences were found in several heart rate variability parameters between the two groups. Additionally, significant differences in heart rate variability parameters were found in swine within 1 hour of inducing hemorrhage in those with traumatic brain injury versus those without. These results were confirmed with stochastic analysis and machine learning was used to generate a model which determined the presence of traumatic brain injury in the setting of hemorrhage shock with 91.6% accuracy.

CONCLUSIONS

 Heart rate variability represents a promising diagnostic tool to aid in the diagnosis of traumatic brain injury within 1 hour of injury.

Modified Early Warning Score vs Cardiac Arrest Risk Triage Score for Prediction of Cardiopulmonary Arrest: A Case-Control Study

Background

Delayed transfer to the intensive care unit (ICU) contributes to increased mortality. Clinical tools, developed to shorten this delay, are especially useful in hospitals where the ideal healthcare provider-to-patient ratio is not met. This study aimed to validate and compare the accuracy of the well-accepted modified early warning score (MEWS) and the newer cardiac arrest risk triage (CART) score in the Philippine setting.

Patients and methods

This case-control study involved 82 adult patients admitted to the Philippine Heart Center. Patients who had cardiopulmonary (CP) arrest at the wards and those transferred to the ICU were included. Vital signs and alert-verbal-pain-unresponsive (AVPU) scales were recorded from recruitment until 48 hours prior to CP arrest or ICU transfer. The MEWS and CART scores were computed at specific time points and compared using measures of validity.

Results

The highest accuracy was obtained by the CART score with a cut-off of ≥12 at 8 hours prior to CP arrest or ICU transfer, with a specificity of 80.43% and sensitivity of 66.67%. At this time point, the MEWS with a cut-off of ≥3 had a specificity of 78.26% but a lower sensitivity of 58.33%. The area under the curve (AUC) analysis revealed that these differences were not statistically significant.

Conclusion

We recommend an MEWS threshold of 3 and a CART score threshold of 12 to help identify patients at risk for clinical deterioration. The CART score had comparable accuracy to the MEWS, but the latter's computation may be easier.

How to cite this article

Tan ADA, Permejo CC, Torres MCD. Modified Early Warning Score vs Cardiac Arrest Risk Triage Score for Prediction of Cardiopulmonary Arrest: A Case-Control Study. Indian J Crit Care Med 2022;26(7):780-785.

Machine Learning-Based Cardiac Arrest Prediction for Early Warning System

The early warning system detects early and responds quickly to emergencies in high-risk patients, such as cardiac arrest in hospitalized patients. However, traditional early warning systems have the problem of frequent false alarms due to low positive predictive value and sensitivity. We conducted early prediction research on cardiac arrest using time-series data such as biosignal and laboratory data. To derive the data attributes that affect the occurrence of cardiac arrest, we performed a correlation analysis between the occurrence of cardiac arrest and the biosignal data and laboratory data. To improve the positive predictive value and sensitivity of early cardiac arrest prediction, we evaluated the performance according to the length of the time series of measured biosignal data, laboratory data, and patient data range. We propose a machine learning and deep learning algorithm: the decision tree, random forest, logistic regression, long short-term memory (LSTM), gated recurrent unit (GRU) model, and the LSTM–GRU hybrid model. We evaluated cardiac arrest prediction models. In the case of our proposed LSTM model, the positive predictive value was 85.92% and the sensitivity was 89.70%.

Fully implantable wireless batteryless vascular electronics with printed soft sensors for multiplex sensing of hemodynamics

The continuous monitoring of hemodynamics attainable with wireless implantable devices would improve the treatment of vascular diseases. However, demanding requirements of size, wireless operation, and compatibility with endovascular procedures have limited the development of vascular electronics. Here, we report an implantable, wireless vascular electronic system, consisting of a multimaterial inductive stent and printed soft sensors capable of real-time monitoring of arterial pressure, pulse rate, and flow without batteries or circuits. Developments in stent design achieve an enhanced wireless platform while matching conventional stent mechanics. The fully printed pressure sensors demonstrate fast response times, high durability, and sensing at small bending radii. The device is monitored via inductive coupling at communication distances notably larger than prior vascular sensors. The wireless electronic system is validated in artery models, while minimally invasive catheter implantation is demonstrated in an in vivo rabbit study. Overall, the vascular system offers an adaptable framework for comprehensive monitoring of hemodynamics.

Intelligent monitoring of noxious stimulation during anaesthesia based on heart rate variability analysis

Research based on medical signals has received significant attention in recent years. If the patients' states can be accurately monitored based on medical signals, it greatly benefits both doctors and patients. This paper proposes a method to extract signal features from heart rate variability signals and classify patients' states using the long short-term memory network and enable effective monitoring of noxious stimulation. For data processing, the heart rate variability signal is decomposed and recombined by the empirical mode decomposition method, and the signal features of the noxious stimulation are extracted by the sliding time window method. Compared with the average accuracy of direct classifications, the classification accuracy based on the proposed method is proved more accurate. The model based on the extracted features proposed can realize the classification of consciousness and general anaesthesia with an accuracy rate of more than 90% and accurately estimate the occurrence of tracheal intubation stimulation. Furthermore, this study shows that combining the deep learning neural network with the extracted more effective signal features under different states and stresses can classify the states with high accuracy. Therefore, it is promising to apply the deep learning method in researching the autonomic nervous system.

Machine learning-based prediction of critical illness in children visiting the emergency department

OBJECTIVES

Triage is an essential emergency department (ED) process designed to provide timely management depending on acuity and severity; however, the process may be inconsistent with clinical and hospitalization outcomes. Therefore, studies have attempted to augment this process with machine learning models, showing advantages in predicting critical conditions and hospitalization outcomes. The aim of this study was to utilize nationwide registry data to develop a machine learning-based classification model to predict the clinical course of pediatric ED visits.

METHODS

This cross-sectional observational study used data from the National Emergency Department Information System on emergency visits of children under 15 years of age from January 1, 2016, to December 31, 2017. The primary and secondary outcomes were to identify critically ill children and predict hospitalization from triage data, respectively. We developed and tested a random forest model with the under sampled dataset and validated the model using the entire dataset. We compared the model's performance with that of the conventional triage system.

RESULTS

A total of 2,621,710 children were eligible for the analysis and included 12,951 (0.5%) critical outcomes and 303,808 (11.6%) hospitalizations. After validation, the area under the receiver operating characteristic curve was 0.991 (95% confidence interval [CI] 0.991-0.992) for critical outcomes and 0.943 (95% CI 0.943-0.944) for hospitalization, which were higher than those of the conventional triage system.

CONCLUSIONS

The machine learning-based model using structured triage data from a nationwide database can effectively predict critical illness and hospitalizations among children visiting the ED.

Remote Cardiac Monitoring in Patients With Heart Failure: A Review

Importance

Heart failure (HF) is often characterized by an insidious disease course leading to frequent rehospitalizations and a high use of ambulatory care. Remote cardiac monitoring is a promising approach to detect worsening HF early and intervene prior to an overt decompensation.

Observations

Recently, a multitude of novel technologies for remote cardiac monitoring (RCM) in patients with HF have been developed and are undergoing clinical trials. This development has been accelerated by the COVID-19 pandemic.

Conclusions and Relevance

This review summarizes the major clinical trials on RCM in patients with HF and present the most recent developments in noninvasive and invasive RCM technologies.

Advances in Cardiac Pacing: Arrhythmia Prediction, Prevention and Control Strategies

Cardiac arrhythmias constitute a tremendous burden on healthcare and are the leading cause of mortality worldwide. An alarming number of people have been reported to manifest sudden cardiac death as the first symptom of cardiac arrhythmias, accounting for about 20% of all deaths annually. Furthermore, patients prone to atrial tachyarrhythmias such as atrial flutter and fibrillation often have associated comorbidities including hypertension, ischemic heart disease, valvular cardiomyopathy and increased risk of stroke. Technological advances in electrical stimulation and sensing modalities have led to the proliferation of medical devices including pacemakers and implantable defibrillators, aiming to restore normal cardiac rhythm. However, given the complex spatiotemporal dynamics and non-linearity of the human heart, predicting the onset of arrhythmias and preventing the transition from steady state to unstable rhythms has been an extremely challenging task. Defibrillatory shocks still remain the primary clinical intervention for lethal ventricular arrhythmias, yet patients with implantable cardioverter defibrillators often suffer from inappropriate shocks due to false positives and reduced quality of life. Here, we aim to present a comprehensive review of the current advances in cardiac arrhythmia prediction, prevention and control strategies. We provide an overview of traditional clinical arrhythmia management methods and describe promising potential pacing techniques for predicting the onset of abnormal rhythms and effectively suppressing cardiac arrhythmias. We also offer a clinical perspective on bridging the gap between basic and clinical science that would aid in the assimilation of promising anti-arrhythmic pacing strategies.

Accuracy of Machine Learning Models to Predict In-hospital Cardiac Arrest: A Systematic Review

PURPOSE/AIMS

Despite advances in healthcare, the incidence of in-hospital cardiac arrest (IHCA) has continued to rise for the past decade. Identifying those patients at risk has proven challenging. Our objective was to conduct a systematic review of the literature to compare the IHCA predictive performance of machine learning (ML) models with the Modified Early Warning Score (MEWS).

DESIGN

The systematic review was conducted following the Preferred Reporting Items of Systematic Review and Meta-Analysis guidelines and registered on PROSPERO CRD42020182357.

METHOD

Data extraction was completed using the Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modeling Studies checklist. The risk of bias and applicability were evaluated using the Prediction model Risk of Bias Assessment Tool.

RESULTS

Nine articles were included in this review that developed or validated IHCA prediction models and compared them with the MEWS. The studies by Jang et al and Kim et al showed that their ML models outperformed MEWS to predict IHCA with good to excellent predictive performance.

CONCLUSIONS

The ML models presented in this systematic review demonstrate a novel approach to predicting IHCA. All included studies suggest that ML models had similar or better predictive performance compared with MEWS. However, there is substantial variability in performance measures and concerns for risk of bias.

A machine learning model to predict critical care outcomes in patient with chest pain visiting the emergency department

Background

Currently, the risk stratification of critically ill patient with chest pain is a challenge. We aimed to use machine learning approach to predict the critical care outcomes in patients with chest pain, and simultaneously compare its performance with HEART, GRACE, and TIMI scores.

Methods

This was a retrospective, case-control study in patients with acute non-traumatic chest pain who presented to the emergency department (ED) between January 2017 and December 2019. The outcomes included cardiac arrest, transfer to ICU, and death during treatment in ED. In the randomly sampled training set (70%), a LASSO regression model was developed, and presented with nomogram. The performance was measured in both training set (70% participants) and testing set (30% participants), and findings were compared with the three widely used scores.

Results

We proposed a LASSO regression model incorporating mode of arrival, reperfusion therapy, Killip class, systolic BP, serum creatinine, creatine kinase-MB, and brain natriuretic peptide as independent predictors of critical care outcomes in patients with chest pain. Our model significantly outperformed the HEART, GRACE, TIMI score with AUC of 0.953 (95%CI: 0.922–0.984), 0.754 (95%CI: 0.675–0.832), 0.747 (95%CI: 0.664–0.829), 0.735 (95%CI: 0.655–0.815), respectively. Consistently, our model demonstrated better outcomes regarding the metrics of accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and F1 score. Similarly, the decision curve analysis elucidated a greater net benefit of our model over the full ranges of clinical thresholds.

Conclusion

We present an accurate model for predicting the critical care outcomes in patients with chest pain, and provide substantial support to its application as a decision-making tool in ED.

Predicting in-hospital mortality in adult non-traumatic emergency department patients: a retrospective comparison of the Modified Early Warning Score (MEWS) and machine learning approach

Background

A feasible and accurate risk prediction systems for emergency department (ED) patients is urgently required. The Modified Early Warning Score (MEWS) is a wide-used tool to predict clinical outcomes in ED. Literatures showed that machine learning (ML) had better predictability in specific patient population than traditional scoring system. By analyzing a large multicenter dataset, we aim to develop a ML model to predict in-hospital morality of the adult non traumatic ED patients for different time stages, and comparing performance with other ML models and MEWS.

Methods

A retrospective observational cohort study was conducted in five Taiwan EDs including two tertiary medical centers and three regional hospitals. All consecutively adult (>17 years old) non-traumatic patients admit to ED during a 9-year period (January first, 2008 to December 31th, 2016) were included. Exclusion criteria including patients with (1) out-of-hospital cardiac arrest and (2) discharge against medical advice and transferred to other hospital (3) missing collect variables. The primary outcome was in-hospital mortality and were categorized into 6, 24, 72, 168 hours mortality. MEWS was calculated by systolic blood pressure, pulse rate, respiratory rate, body temperature, and level of consciousness. An ensemble supervised stacking ML model was developed and compared to sensitive and unsensitive Xgboost, Random Forest, and Adaboost. We conducted a performance test and examine both the area under the receiver operating characteristic (AUROC) and the area under the precision and recall curve (AUPRC) as the comparative measures.

Result

After excluding 182,001 visits (7.46%), study group was consisted of 24,37,326 ED visits. The dataset was split into 67% training data and 33% test data for ML model development. There was no statistically difference found in the characteristics between two groups. For the prediction of 6, 24, 72, 168 hours in-hospital mortality, the AUROC of MEW and ML mode was 0.897, 0.865, 0.841, 0.816 and 0.939, 0.928, 0.913, 0.902 respectively. The stacking ML model outperform other ML model as well. For the prediction of in-hospital mortality over 48-hours, AUPRC performance of MEWS drop below 0.1, while the AUPRC of ML mode was 0.317 in 6 hours and 0.2150 in 168 hours. For each time frame, ML model achieved statistically significant higher AUROC and AUPRC than MEWS (all P < 0.001). Both models showed decreasing prediction ability as time elapse, but there was a trend that the gap of AUROC values between two model increases gradually (P < 0.001). Three MEWS thresholds (score >3, >4, and >5) were determined as baselines for comparison, ML mode consistently showed improved or equally performance in sensitivity, PPV, NPV, but not in specific.

Conclusion

Stacking ML methods improve predicted in-hospital mortality than MEWS in adult non-traumatic ED patients, especially in the prediction of delayed mortality.

A review on utilizing machine learning technology in the fields of electronic emergency triage and patient priority systems in telemedicine: Coherent taxonomy, motivations, open research challenges and recommendations for intelligent future work

BACKGROUND

With the remarkable increasing in the numbers of patients, the triaging and prioritizing patients into multi-emergency level is required to accommodate all the patients, save more lives, and manage the medical resources effectively. Triaging and prioritizing patients becomes particularly challenging especially for the patients who are far from hospital and use telemedicine system. To this end, the researchers exploiting the useful tool of machine learning to address this challenge. Hence, carrying out an intensive investigation and in-depth study in the field of using machine learning in E-triage and patient priority are essential and required.

OBJECTIVES

This research aims to (1) provide a literature review and an in-depth study on the roles of machine learning in the fields of electronic emergency triage (E-triage) and prioritize patients for fast healthcare services in telemedicine applications. (2) highlight the effectiveness of machine learning methods in terms of algorithms, medical input data, output results, and machine learning goals in remote healthcare telemedicine systems. (3) present the relationship between machine learning goals and the electronic triage processes specifically on the: triage levels, medical features for input, outcome results as outputs, and the relevant diseases. (4), the outcomes of our analyses are subjected to organize and propose a cross-over taxonomy between machine learning algorithms and telemedicine structure. (5) present lists of motivations, open research challenges and recommendations for future intelligent work for both academic and industrial sectors in telemedicine and remote healthcare applications.

METHODS

An intensive research is carried out by reviewing all articles related to the field of E-triage and remote priority systems that utilise machine learning algorithms and sensors. We have searched all related keywords to investigate the databases of Science Direct, IEEE Xplore, Web of Science, PubMed, and Medline for the articles, which have been published from January 2012 up to date.

RESULTS

A new crossover matching between machine learning methods and telemedicine taxonomy is proposed. The crossover-taxonomy is developed in this study to identify the relationship between machine learning algorithm and the equivalent telemedicine categories whereas the machine learning algorithm has been utilized. The impact of utilizing machine learning is composed in proposing the telemedicine architecture based on synchronous (real-time/ online) and asynchronous (store-and-forward / offline) structure. In addition to that, list of machine learning algorithms, list of the performance metrics, list of inputs data and outputs results are presented. Moreover, open research challenges, the benefits of utilizing machine learning and the recommendations for new research opportunities that need to be addressed for the synergistic integration of multidisciplinary works are organized and presented accordingly.

DISCUSSION

The state-of-the-art studies on the E-triage and priority systems that utilise machine learning algorithms in telemedicine architecture are discussed. This approach allows the researchers to understand the modernisation of healthcare systems and the efficient use of artificial intelligence and machine learning. In particular, the growing worldwide population and various chronic diseases such as heart chronic diseases, blood pressure and diabetes, require smart health monitoring systems in E-triage and priority systems, in which machine learning algorithms could be greatly beneficial.

CONCLUSIONS

Although research directions on E-triage and priority systems that use machine learning algorithms in telemedicine vary, they are equally essential and should be considered. Hence, we provide a comprehensive review to emphasise the advantages of the existing research in multidisciplinary works of artificial intelligence, machine learning and healthcare services.

An Artificial Neural Network-Based Pediatric Mortality Risk Score: Development and Performance Evaluation Using Data From a Large North American Registry

Background

In the pediatric intensive care unit (PICU), quantifying illness severity can be guided by risk models to enable timely identification and appropriate intervention. Logistic regression models, including the pediatric index of mortality 2 (PIM-2) and pediatric risk of mortality III (PRISM-III), produce a mortality risk score using data that are routinely available at PICU admission. Artificial neural networks (ANNs) outperform regression models in some medical fields.

Objective

In light of this potential, we aim to examine ANN performance, compared to that of logistic regression, for mortality risk estimation in the PICU.

Methods

The analyzed data set included patients from North American PICUs whose discharge diagnostic codes indicated evidence of infection and included the data used for the PIM-2 and PRISM-III calculations and their corresponding scores. We stratified the data set into training and test sets, with approximately equal mortality rates, in an effort to replicate real-world data. Data preprocessing included imputing missing data through simple substitution and normalizing data into binary variables using PRISM-III thresholds. A 2-layer ANN model was built to predict pediatric mortality, along with a simple logistic regression model for comparison. Both models used the same features required by PIM-2 and PRISM-III. Alternative ANN models using single-layer or unnormalized data were also evaluated. Model performance was compared using the area under the receiver operating characteristic curve (AUROC) and the area under the precision recall curve (AUPRC) and their empirical 95% CIs.

Results

Data from 102,945 patients (including 4068 deaths) were included in the analysis. The highest performing ANN (AUROC 0.871, 95% CI 0.862-0.880; AUPRC 0.372, 95% CI 0.345-0.396) that used normalized data performed better than PIM-2 (AUROC 0.805, 95% CI 0.801-0.816; AUPRC 0.234, 95% CI 0.213-0.255) and PRISM-III (AUROC 0.844, 95% CI 0.841-0.855; AUPRC 0.348, 95% CI 0.322-0.367). The performance of this ANN was also significantly better than that of the logistic regression model (AUROC 0.862, 95% CI 0.852-0.872; AUPRC 0.329, 95% CI 0.304-0.351). The performance of the ANN that used unnormalized data (AUROC 0.865, 95% CI 0.856-0.874) was slightly inferior to our highest performing ANN; the single-layer ANN architecture performed poorly and was not investigated further.

Conclusions

A simple ANN model performed slightly better than the benchmark PIM-2 and PRISM-III scores and a traditional logistic regression model trained on the same data set. The small performance gains achieved by this two-layer ANN model may not offer clinically significant improvement; however, further research with other or more sophisticated model designs and better imputation of missing data may be warranted.

A signal quality assessment-based ECG waveform delineation method used for wearable monitoring systems

Identifying transient and nonpersistent abnormal electrocardiogram (ECG) waveforms by continuously monitoring the high-risk populations is of great importance for the diagnosis, treatment, and prevention of cardiovascular diseases. In recent years, fabric electrodes have been widely used in wearable devices because of their non-irritating properties and better comfort than traditional AgCl electrodes. However, the motion noise caused by the relative movement between the fabric electrodes and skin affects the quality of ECGs and reduces the accuracy of diagnosis. Therefore, delineating the ECG waveforms that are recorded from wearable devices with varying levels of noise is still challenging. In this study, a signal quality assessment (SQA)-based ECG waveform delineation method that is used for wearable systems was developed. The ECG signal was first preprocessed by a bandpass filter. Five indices, including the multiscale nonlinear amplitude statistical distribution (adSQI₁, adSQI₂), the proportion of energy-related to T wave (ptSQI), and heart rates computed from R waves and T waves (rHR and tHR, respectively), were then calculated from the preprocessed ECG signal. The signals were classified as good, acceptable, and unacceptable ECGs by combining these indices through the use of a neural network. Subsequently, the R waves or/and T waves were identified for the corresponding feature interpretations based on the SQA results. ECGs that were recorded from the chest belts from 29 volunteers at different activity statuses were divided into 4-s segments. A total of 7133 manually labeled segments were used to derive (4985 segments) and validate (2148 segments) the algorithm. The adSQI₁, adSQI₂, tHR, and rHR characteristics were significantly different among the good, acceptable, and unacceptable ECGs. The ptSQI value was considerably higher in the good ECGs than in the acceptable and unacceptable ECGs. The ECG segments of different quality levels were classified with an accuracy of 96.74% by using the proposed SQA method. The R waves and T waves were identified with accuracies of 99.95% and 99.57%, respectively, for segments that were classified as acceptable and/or good. The SQA-based ECG waveform delineation method can perform reliable analysis and has the potential to be applied in wearable ECG systems for the early diagnosis and prevention of cardiovascular diseases.

Development and Validation of Unplanned Extubation Prediction Models Using Intensive Care Unit Data: Retrospective, Comparative, Machine Learning Study

Background

Patient safety in the intensive care unit (ICU) is one of the most critical issues, and unplanned extubation (UE) is considered the most adverse event for patient safety. Prevention and early detection of such an event is an essential but difficult component of quality care.

Objective

This study aimed to develop and validate prediction models for UE in ICU patients using machine learning.

Methods

This study was conducted in an academic tertiary hospital in Seoul, Republic of Korea. The hospital had approximately 2000 inpatient beds and 120 ICU beds. As of January 2019, the hospital had approximately 9000 outpatients on a daily basis. The number of annual ICU admissions was approximately 10,000. We conducted a retrospective study between January 1, 2010, and December 31, 2018. A total of 6914 extubation cases were included. We developed a UE prediction model using machine learning algorithms, which included random forest (RF), logistic regression (LR), artificial neural network (ANN), and support vector machine (SVM). For evaluating the model’s performance, we used the area under the receiver operating characteristic curve (AUROC). The sensitivity, specificity, positive predictive value, negative predictive value, and F1 score were also determined for each model. For performance evaluation, we also used a calibration curve, the Brier score, and the integrated calibration index (ICI) to compare different models. The potential clinical usefulness of the best model at the best threshold was assessed through a net benefit approach using a decision curve.

Results

Among the 6914 extubation cases, 248 underwent UE. In the UE group, there were more males than females, higher use of physical restraints, and fewer surgeries. The incidence of UE was higher during the night shift as compared to the planned extubation group. The rate of reintubation within 24 hours and hospital mortality were higher in the UE group. The UE prediction algorithm was developed, and the AUROC for RF was 0.787, for LR was 0.762, for ANN was 0.763, and for SVM was 0.740.

Conclusions

We successfully developed and validated machine learning–based prediction models to predict UE in ICU patients using electronic health record data. The best AUROC was 0.787 and the sensitivity was 0.949, which was obtained using the RF algorithm. The RF model was well-calibrated, and the Brier score and ICI were 0.129 and 0.048, respectively. The proposed prediction model uses widely available variables to limit the additional workload on the clinician. Further, this evaluation suggests that the model holds potential for clinical usefulness.

Prediction of In-Hospital Cardiac Arrest Using Shallow and Deep Learning

Sudden cardiac arrest can leave serious brain damage or lead to death, so it is very important to predict before a cardiac arrest occurs. However, early warning score systems including the National Early Warning Score, are associated with low sensitivity and false positives. We applied shallow and deep learning to predict cardiac arrest to overcome these limitations. We evaluated the performance of the Synthetic Minority Oversampling Technique Ratio. We evaluated the performance using a Decision Tree, a Random Forest, Logistic Regression, Long Short-Term Memory model, Gated Recurrent Unit model, and LSTM–GRU hybrid models. Our proposed Logistic Regression demonstrated a higher positive predictive value and sensitivity than traditional early warning systems.

The EMS Modified Early Warning Score (EMEWS): A Simple Count of Vital Signs as a Predictor of Out-of-Hospital Cardiac Arrests

Objective: For patients at risk for out-of-hospital cardiac arrest (OHCA) after Emergency Medical Services (EMS) arrival, outcomes may be mitigated by identifying impending arrests and intervening before they occur. Tools such as the Modified Early Warning Score (MEWS) have been developed to determine the risk of arrest, but involve relatively complicated algorithms that can be impractical to compute in the prehospital environment. A simple count of abnormal vital signs, the "EMS Modified Early Warning Score" (EMEWS), may represent a more practical alternative. We sought to compare to the ability of MEWS and EMEWS to identify patients at risk for EMS-witnessed OHCA.Methods: We conducted a retrospect analysis of the 2018 ESO Data Collaborative database of EMS encounters. Patients without cardiac arrest before EMS arrival were categorized into those who did or did not have an EMS-witnessed arrest. MEWS was evaluated without its temperature component (MEWS-T). The performance of MEWS-T and EMEWS in predicting EMS witnessed arrest was evaluated by comparing receiver-operating characteristic curves.Results: Of 369,064 included encounters, 4,651 were EMS witnessed arrests. MEWS-T demonstrated an area under the curve (AUC) of 0.79 (95% CI: 0.79 - 0.80), with 86.8% sensitivity and 51.0% specificity for MEWS-T ≥ 3. EMEWS demonstrated an AUC of 0.74 (95% CI: 0.73 - 0.75), with 81.3% sensitivity and 53.9% specificity for EMEWS ≥ 2.Conclusions: EMEWS showed a similar ability to predict EMS-witnessed cardiac arrest compared to MEWS-T, despite being significantly simpler to compute. Further study is needed to evaluate whether the implementation of EMEWS can aid EMS clinicians in anticipating and preventing OHCA.

Machine learning in patient flow: a review

This work is a review of the ways in which machine learning has been used in order to plan, improve or aid the problem of moving patients through healthcare services. We decompose the patient flow problem into four subcategories: prediction of demand on a healthcare institution, prediction of the demand and resource required to transfer patients from the emergency department to the hospital, prediction of potential resource required for the treatment and movement of inpatients and prediction of length-of-stay and discharge timing. We argue that there are benefits to both approaches of considering the healthcare institution as a whole as well as the patient by patient case and that ideally a combination of these would be best for improving patient flow through hospitals. We also argue that it is essential for there to be a shared dataset that will allow researchers to benchmark their algorithms on and thereby allow future researchers to build on that which has already been done. We conclude that machine learning for the improvement of patient flow is still a young field with very few papers tailor-making machine learning methods for the problem being considered. Future works should consider the need to transfer algorithms trained on a dataset to multiple hospitals and allowing for dynamic algorithms which will allow real-time decision-making to help clinical staff on the shop floor.

Predictive value of Modified Early Warning Score (MEWS) and Revised Trauma Score (RTS) for the short-term prognosis of emergency trauma patients: a retrospective study

OBJECTIVES

This study aimed to assess the predictive value of the Modified Early Warning Score (MEWS) and Revised Trauma Score (RTS) for emergency trauma patients who died within 24 hours.

DESIGN

A retrospective, single-centred study.

SETTING

This study was conducted at a tertiary hospital in Southern China.

PARTICIPANTS

A total of 1739 patients with acute trauma, aged 16 years or older who presented to the emergency department from 1 November 2016 to 30 November 2019, were included.

INTERVENTIONS NONE

None.

OUTCOME

24-hour mortality was the primary outcome of trauma.

RESULTS

1739 patients were divided into the survival group (1709 patients,98.27%), and the non-survival group (30 patients,1.73%). Crude OR and adjusted OR of MEWS were 1.99, 95% CI (1.73 to 2.29), and 2.00, 95% CI (1.74 to 2.31), p<0.001, respectively. Crude OR and adjusted OR of RTS were 0.62, 95% CI (0.55 to 0.69) and 0.61, 95% CI (0.55 to 0.68), p<0.001, respectively. The area under the curve of MEWS was significantly higher than that of RTS (p=0.005): 0.927, 95% CI (0.914 to 0.939) vs 0.799, 95% CI (0.779 to 0.817).

CONCLUSIONS

Both MEWS and RTS were independent predictors of the short-term prognosis in emergency trauma patients, MEWS had better predictive efficacy.

Consensus on predictors of clinical deterioration in emergency departments: A Delphi process study

AIM

The study aim was to determine relevance and applicability of generic predictors of clinical deterioration in emergency departments based on consensus among clinicians.

METHODS

Thirty-three predictors of clinical deterioration identified from literature were assessed in a modified two-stage Delphi-process. Sixty-eight clinicians (physicians and nurses) participated in the first round and 48 in the second round; all treating hospitalized patients in Danish emergency departments, some with pre-hospital experience. The panel rated the predictors for relevance (relevant marker of clinical deterioration) and applicability (change in clinical presentation over time, generic in nature and possible to detect bedside). They rated their level of agreement on a 9-point Likert scale and were also invited to propose additional generic predictors between the rounds. New predictors suggested by more than one clinician were included in the second round along with non-consensus predictors from the first round. Final decisions of non-consensus predictors after second round were made by a research group and an impartial physician.

RESULTS

The Delphi-process resulted in 19 clinically relevant and applicable predictors based on vital signs and parameters (respiratory rate, saturation, dyspnoea, systolic blood pressure, pulse rate, abnormal electrocardiogram, altered mental state and temperature), biochemical tests (serum c-reactive protein, serum bicarbonate, serum lactate, serum pH, serum potassium, glucose, leucocyte counts and serum haemoglobin), objective clinical observations (skin conditions) and subjective clinical observations (pain reported as new or escalating, and relatives' concerns).

CONCLUSION

The Delphi-process led to consensus of 19 potential predictors of clinical deterioration widely accepted as relevant and applicable in emergency departments.

Machine Learning Versus Usual Care for Diagnostic and Prognostic Prediction in the Emergency Department: A Systematic Review

OBJECTIVE

Having shown promise in other medical fields, we sought to determine whether machine learning (ML) models perform better than usual care in diagnostic and prognostic prediction for emergency department (ED) patients.

METHODS

In this systematic review, we searched MEDLINE, Embase, Central, and CINAHL from inception to October 17, 2019. We included studies comparing diagnostic and prognostic prediction of ED patients by ML models to usual care methods (triage-based scores, clinical prediction tools, clinician judgment) using predictor variables readily available to ED clinicians. We extracted commonly reported performance metrics of model discrimination and classification. We used the PROBAST tool for risk of bias assessment (PROSPERO registration: CRD42020158129).

RESULTS

The search yielded 1,656 unique records, of which 23 studies involving 16,274,647 patients were included. In all seven diagnostic studies, ML models outperformed usual care in all performance metrics. In six studies assessing in-hospital mortality, the best-performing ML models had better discrimination (area under the receiver operating characteristic curve [AUROC] =0.74-0.94) than any clinical decision tool (AUROC =0.68-0.81). In four studies assessing hospitalization, ML models had better discrimination (AUROC =0.80-0.83) than triage-based scores (AUROC =0.68-0.82). Clinical heterogeneity precluded meta-analysis. Most studies had high risk of bias due to lack of external validation, low event rates, and insufficient reporting of calibration.

CONCLUSIONS

Our review suggests that ML may have better prediction performance than usual care for ED patients with a variety of clinical presentations and outcomes. However, prediction model reporting guidelines should be followed to provide clinically applicable data. Interventional trials are needed to assess the impact of ML models on patient-centered outcomes.

Machine Learning for Clinical Outcome Prediction

Clinical decision-making in healthcare is already being influenced by predictions or recommendations made by data-driven machines. Numerous machine learning applications have appeared in the latest clinical literature, especially for outcome prediction models, with outcomes ranging from mortality and cardiac arrest to acute kidney injury and arrhythmia. In this review article, we summarize the state-of-the-art in related works covering data processing, inference, and model evaluation, in the context of outcome prediction models developed using data extracted from electronic health records. We also discuss limitations of prominent modeling assumptions and highlight opportunities for future research.

Prognostic machine learning models for COVID-19 to facilitate decision making

An increasing number of COVID-19 cases worldwide has overwhelmed the healthcare system. Physicians are struggling to allocate resources and to focus their attention on high-risk patients, partly because early identification of high-risk individuals is difficult. This can be attributed to the fact that COVID-19 is a novel disease and its pathogenesis is still partially understood. However, machine learning algorithms have the capability to analyse a large number of parameters within a short period of time to identify the predictors of disease outcome. Implementing such an algorithm to predict high-risk individuals during the early stages of infection would be helpful in decision making for clinicians such that irreversible damage could be prevented. Here, we propose recommendations to develop prognostic machine learning models using electronic health records so that a real-time risk score can be developed for COVID-19.

Integration of novel monitoring devices with machine learning technology for scalable cardiovascular management

Ambulatory monitoring is increasingly important for cardiovascular care but is often limited by the unpredictability of cardiovascular events, the intermittent nature of ambulatory monitors and the variable clinical significance of recorded data in patients. Technological advances in computing have led to the introduction of novel physiological biosignals that can increase the frequency at which abnormalities in cardiovascular parameters can be detected, making expert-level, automated diagnosis a reality. However, use of these biosignals for diagnosis also raises numerous concerns related to accuracy and actionability within clinical guidelines, in addition to medico-legal and ethical issues. Analytical methods such as machine learning can potentially increase the accuracy and improve the actionability of device-based diagnoses. Coupled with interoperability of data to widen access to all stakeholders, seamless connectivity (an internet of things) and maintenance of anonymity, this approach could ultimately facilitate near-real-time diagnosis and therapy. These tools are increasingly recognized by regulatory agencies and professional medical societies, but several technical and ethical issues remain. In this Review, we describe the current state of cardiovascular monitoring along the continuum from biosignal acquisition to the identification of novel biosensors and the development of analytical techniques and ultimately to regulatory and ethical issues. Furthermore, we outline new paradigms for cardiovascular monitoring.

Advances in cardiovascular monitoring technologies have resulted in an influx of consumer-targeted wearable sensors that have the potential to detect numerous heart conditions. In this Review, Krittanawong and colleagues describe processes involved in biosignal acquisition and analysis of cardiovascular monitors, as well as their associated ethical, regulatory and legal challenges.

Advances in the use of cardiovascular monitoring technologies, such as the development of novel portable sensors and machine learning algorithms that can provide near-real-time diagnosis, have the potential to provide personalized care.

Wearable sensor technologies can detect numerous biosignals, such as cardiac output, blood-pressure levels and heart rhythm, and can integrate multiple modalities.

The use of novel biosignals for diagnosis raises concerns regarding accuracy and actionability within clinical guidelines, in addition to medical, legal and ethical issues.

Machine learning-based interpretation of biosensor data can facilitate rapid evaluation of the haemodynamic consequences of heart failure or arrhythmias, but is limited by the presence of noise and training data that might not be representative of the real-world clinical setting.

The use of data derived from cardiovascular monitoring devices is associated with numerous challenges, such as data security, accessibility and ownership, in addition to other ethical and regulatory concerns.

Prediction of Sudden Cardiac Death Risk with a Support Vector Machine Based on Heart Rate Variability and Heartprint Indices

Most methods for sudden cardiac death (SCD) prediction require long-term (24 h) electrocardiogram recordings to measure heart rate variability (HRV) indices or premature ventricular complex indices (with the heartprint method). This work aimed to identify the best combinations of HRV and heartprint indices for predicting SCD based on short-term recordings (1000 heartbeats) through a support vector machine (SVM). Eleven HRV indices and five heartprint indices were measured in 135 pairs of recordings (one before an SCD episode and another without SCD as control). SVMs (defined with a radial basis function kernel with hyperparameter optimization) were trained with this dataset to identify the 13 best combinations of indices systematically. Through 10-fold cross-validation, the best area under the curve (AUC) value as a function of γ (gamma) and cost was identified. The predictive value of the identified combinations had AUCs between 0.80 and 0.86 and accuracies between 80 and 86%. Further SVM performance tests on a different dataset of 68 recordings (33 before SCD and 35 as control) showed AUC = 0.68 and accuracy = 67% for the best combination. The developed SVM may be useful for preventing imminent SCD through early warning based on electrocardiogram (ECG) or heart rate monitoring.

Predicting cardiac arrest in the emergency department

In-hospital cardiac arrest remains a leading cause of death: roughly 300,000 in-hospital cardiac arrests occur each year in the United States, ≈10% of which occur in the emergency department. ED-based cardiac arrest may represent a subset of in-hospital cardiac arrest with a higher proportion of reversible etiologies and a higher potential for neurologically intact survival. Patients presenting to the ED have become increasingly complex, have a high burden of critical illness, and face crowded departments with thinly stretched resources. As a result, patients in the ED are vulnerable to unrecognized clinical deterioration that may lead to ED-based cardiac arrest. Efforts to identify patients who may progress to ED-based cardiac arrest have traditionally been approached through identification of critically ill patients at triage and the identification of patients who unexpectedly deteriorate during their stay in the ED. Interventions to facilitate appropriate triage and resource allocation, as well as earlier identification of patients at risk of deterioration in the ED, could potentially allow for both prevention of cardiac arrest and optimization of outcomes from ED-based cardiac arrest. This review will discuss the epidemiology of ED-based cardiac arrest, as well as commonly used approaches to predict ED-based cardiac arrest and highlight areas that require further research to improve outcomes for this population.

Generalizability of SuperAlarm via Cross-Institutional Performance Evaluation

Bedside patient monitors are ubiquitous tools in modern critical care units to provide timely patient status. However, current systems suffer from high volume of false alarms leading to alarm fatigue, one of top technical hazards in clinical settings. Many studies are racing to develop improved algorithms towards precision patient monitoring, while little has been done to investigate the aspect of algorithm generalizability across different health institutions. Our group has been developing an evolving framework termed SuperAlarm that extracts multivariate patterns in data streams (monitor alarms, electronic health records and physiologic waveforms) of modern health enterprise to predict patient deterioration and has demonstrated great potential in mitigating alarm fatigue. In this study, we further investigate the generalizability of SuperAlarm by designing a comprehensive approach to achieve performance comparison in predicting in-hospital code blue (CB) events across two health institutions. SuperAlarm model trained with alarm data in one institution is tested on both internal and external test sets. Results show comparable performance with sensitivity up to 80% within one-hour window of events and over 90% in reduction of false alarms in both institutions. Cross-institutional performance agreement can be further improved by predicting a more stringent CB subtype (cardiopulmonary arrest), with internal sensitivity lying within 95% confident interval of external one up to 8-hour before event onset. The cross-institutional performance comparison offers first-hand knowledge on both advantages and challenges in generalizing a prediction algorithm across different institutions, which hold key information to guide the design of model training and deployment strategy.

Risk of mortality and cardiopulmonary arrest in critical patients presenting to the emergency department using machine learning and natural language processing

Emergency department triage is the first point in time when a patient's acuity level is determined. The time to assign a priority at triage is short and it is vital to accurately stratify patients at this stage, since under-triage can lead to increased morbidity, mortality and costs. Our aim was to present a model that can assist healthcare professionals in triage decision making, namely in the stratification of patients through the risk prediction of a composite critical outcome-mortality and cardiopulmonary arrest. Our study cohort consisted of 235826 adult patients triaged at a Portuguese Emergency Department from 2012 to 2016. Patients were assigned to emergent, very urgent or urgent priorities of the Manchester Triage System (MTS). Demographics, clinical variables routinely collected at triage and the patients' chief complaint were used. Logistic regression, random forests and extreme gradient boosting were developed using all available variables. The term frequency-inverse document frequency (TF-IDF) natural language processing weighting factor was applied to vectorize the chief complaint. Stratified random sampling was used to split the data into train (70%) and test (30%) data sets. Ten-fold cross validation was performed in train to optimize model hyper-parameters. The performance obtained with the best model was compared against the reference model-a regularized logistic regression trained using only triage priorities. Extreme gradient boosting exhibited good calibration properties and yielded areas under the receiver operating characteristic and precision-recall curves of 0.96 (95% CI 0.95-0.97) and 0.31 (95% CI 0.26-0.36), respectively. The predictors ranked with higher importance by this model were the Glasgow coma score, the patients' age, pulse oximetry and arrival mode. Compared to the reference, the extreme gradient boosting model using clinical variables and the chief complaint presented higher recall for patients assigned MTS-3 and can identify those who are at risk of the composite outcome.

Development and Validation of a Quick Sepsis-Related Organ Failure Assessment-Based Machine-Learning Model for Mortality Prediction in Patients with Suspected Infection in the Emergency Department

The quick sepsis-related organ failure assessment (qSOFA) score has been introduced to predict the likelihood of organ dysfunction in patients with suspected infection. We hypothesized that machine-learning models using qSOFA variables for predicting three-day mortality would provide better accuracy than the qSOFA score in the emergency department (ED). Between January 2016 and December 2018, the medical records of patients aged over 18 years with suspected infection were retrospectively obtained from four EDs in Korea. Data from three hospitals (n = 19,353) were used as training-validation datasets and data from one (n = 4234) as the test dataset. Machine-learning algorithms including extreme gradient boosting, light gradient boosting machine, and random forest were used. We assessed the prediction ability of machine-learning models using the area under the receiver operating characteristic (AUROC) curve, and DeLong's test was used to compare AUROCs between the qSOFA scores and qSOFA-based machine-learning models. A total of 447,926 patients visited EDs during the study period. We analyzed 23,587 patients with suspected infection who were admitted to the EDs. The median age of the patients was 63 years (interquartile range: 43-78 years) and in-hospital mortality was 4.0% (n = 941). For predicting three-day mortality among patients with suspected infection in the ED, the AUROC of the qSOFA-based machine-learning model (0.86 [95% CI 0.85-0.87]) for three -day mortality was higher than that of the qSOFA scores (0.78 [95% CI 0.77-0.79], p < 0.001). For predicting three-day mortality in patients with suspected infection in the ED, the qSOFA-based machine-learning model was found to be superior to the conventional qSOFA scores.

Entropy-Based Measures of Hypnopompic Heart Rate Variability Contribute to the Automatic Prediction of Cardiovascular Events

Surges in sympathetic activity should be a major contributor to the frequent occurrence of cardiovascular events towards the end of nocturnal sleep. We aimed to investigate whether the analysis of hypnopompic heart rate variability (HRV) could assist in the prediction of cardiovascular disease (CVD). 2217 baseline CVD-free subjects were identified and divided into CVD group and non-CVD group, according to the presence of CVD during a follow-up visit. HRV measures derived from time domain analysis, frequency domain analysis and nonlinear analysis were employed to characterize cardiac functioning. Machine learning models for both long-term and short-term CVD prediction were then constructed, based on hypnopompic HRV metrics and other typical CVD risk factors. CVD was associated with significant alterations in hypnopompic HRV. An accuracy of 81.4% was achieved in short-term prediction of CVD, demonstrating a 10.7% increase compared with long-term prediction. There was a decline of more than 6% in the predictive performance of short-term CVD outcomes without HRV metrics. The complexity of hypnopompic HRV, measured by entropy-based indices, contributed considerably to the prediction and achieved greater importance in the proposed models than conventional HRV measures. Our findings suggest that Hypnopompic HRV assists the prediction of CVD outcomes, especially the occurrence of CVD event within two years.

Heart Rate Variability as a Biomarker of Neurocardiogenic Injury After Subarachnoid Hemorrhage

BACKGROUND

The objective of this study was to examine whether heart rate variability (HRV) measures can be used to detect neurocardiogenic injury (NCI).

METHODS

Three hundred and twenty-six consecutive admissions with aneurysmal subarachnoid hemorrhage (SAH) met criteria for the study. Of 326 subjects, 56 (17.2%) developed NCI which we defined by wall motion abnormality with ventricular dysfunction on transthoracic echocardiogram or cardiac troponin-I > 0.3 ng/mL without electrocardiogram evidence of coronary artery insufficiency. HRV measures (in time and frequency domains, as well as nonlinear technique of detrended fluctuation analysis) were calculated over the first 48 h. We applied longitudinal multilevel linear regression to characterize the relationship of HRV measures with NCI and examine between-group differences at baseline and over time.

RESULTS

There was decreased vagal activity in NCI subjects with a between-group difference in low/high frequency ratio (β 3.42, SE 0.92, p = 0.0002), with sympathovagal balance in favor of sympathetic nervous activity. All time-domain measures were decreased in SAH subjects with NCI. An ensemble machine learning approach translated these measures into a classification tool that demonstrated good discrimination using the area under the receiver operating characteristic curve (AUROC 0.82), the area under precision recall curve (AUPRC 0.75), and a correct classification rate of 0.81.

CONCLUSIONS

HRV measures are significantly associated with our label of NCI and a machine learning approach using features derived from HRV measures can classify SAH patients that develop NCI.

Developing neural network models for early detection of cardiac arrest in emergency department

BACKGROUND

Automated surveillance for cardiac arrests would be useful in overcrowded emergency departments. The purpose of this study is to develop and test artificial neural network (ANN) classifiers for early detection of patients at risk of cardiac arrest in emergency departments.

METHODS

This is a single-center electronic health record (EHR)-based study. The primary outcome was the development of cardiac arrest within 24 h after prediction. Three ANN models were trained: multilayer perceptron (MLP), long-short-term memory (LSTM), and hybrid. These were compared to other classifiers including the modified early warning score (MEWS), logistic regression, and random forest. We used AUROC, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for the comparison.

RESULTS

During the study period, there were a total of 374,605 ED visits and 2,910,321 patient status updates. The ANN models (MLP, LSTM, and hybrid) achieved higher AUROC (AUROC: 0.929, 0.933, and 0.936; 95% confidential interval: 0.926-0.932, 0.930-0.936, and 0.933-0.939, respectively) compared to the non-ANN models, and the hybrid model exhibited the best performance. The ANN classifiers displayed higher performance in most of the test characteristics when the threshold levels of the classifiers were fixed to display the same positive result as those at the three MEWS thresholds (score ≥ 3, ≥4, and ≥5), and when compared with each other.

CONCLUSIONS

The ANN improves upon MEWS and conventional machine learning algorithms for the prediction of cardiac arrests in emergency departments. The hybrid ANN model utilizing both baseline and sequence information achieved the best performance.

Combining quick sequential organ failure assessment score with heart rate variability may improve predictive ability for mortality in septic patients at the emergency department

Background

Although the quick Sequential Organ Failure Assessment (qSOFA) score was recently introduced to identify patients with suspected infection/sepsis, it has limitations as a predictive tool for adverse outcomes. We hypothesized that combining qSOFA score with heart rate variability (HRV) variables improves predictive ability for mortality in septic patients at the emergency department (ED).

Methods

This was a retrospective study using the electronic medical record of a tertiary care hospital in Singapore between September 2014 and February 2017. All patients aged 21 years or older who were suspected with infection/sepsis in the ED and received electrocardiography monitoring with ZOLL X Series Monitor (ZOLL Medical Corporation, Chelmsford, MA) were included. We fitted a logistic regression model to predict the 30-day mortality using one of the HRV variables selected from one of each three domains those previously reported as strong association with mortality (i.e. standard deviation of NN [SDNN], ratio of low frequency to high frequency power [LF/HF], detrended fluctuation analysis α-2 [DFA α-2]) in addition to the qSOFA score. The predictive accuracy was assessed with other scoring systems (i.e. qSOFA alone, National Early Warning Score, and Modified Early Warning Score) using the area under the receiver operating characteristic curve.

Results

A total of 343 septic patients were included. Non-survivors were significantly older (survivors vs. non-survivors, 65.7 vs. 72.9, p <0.01) and had higher qSOFA (0.8 vs. 1.4, p <0.01) as compared to survivors. There were significant differences in HRV variables between survivors and non-survivors including SDNN (23.7s vs. 31.8s, p = 0.02), LF/HF (2.8 vs. 1.5, p = 0.02), DFA α-2 (1.0 vs. 0.7, P < 0.01). Our prediction model using DFA-α-2 had the highest c-statistic of 0.76 (95% CI, 0.70 to 0.82), followed by qSOFA of 0.68 (95% CI, 0.62 to 0.75), National Early Warning Score at 0.67 (95% CI, 0.61 to 0.74), and Modified Early Warning Score at 0.59 (95% CI, 0.53 to 0.67).

Conclusions

Adding DFA-α-2 to the qSOFA score may improve the accuracy of predicting in-hospital mortality in septic patients who present to the ED. Further multicenter prospective studies are required to confirm our results.

Predicting cardiac arrests in pediatric intensive care units

BACKGROUND

Early identification of children at risk for cardiac arrest would allow for skill training associated with improved outcomes and provides a prevention opportunity.

OBJECTIVE

Develop and assess a predictive model for cardiopulmonary arrest using data available in the first 4 h.

METHODS

Data from PICU patients from 8 institutions included descriptive, severity of illness, cardiac arrest, and outcomes.

RESULTS

Of the 10074 patients, 120 satisfying inclusion criteria sustained a cardiac arrest and 67 (55.9%) died. In univariate analysis, patients with cardiac arrest prior to admission were over 6 times and those with cardiac arrests during the first 4 h were over 50 times more likely to have a subsequent arrest. The multivariate logistic regression model performance was excellent (area under the ROC curve = 0.85 and Hosmer-Lemeshow statistic, p = 0.35). The variables with the highest odds ratio's for sustaining a cardiac arrest in the multivariable model were admission from an inpatient unit (8.23 (CI: 4.35-15.54)), and cardiac arrest in the first 4 h (6.48 (CI: 2.07-20.36). The average risk predicted by the model was highest (11.6%) among children sustaining an arrest during hours >4-12 and continued to be high even for days after the risk assessment period; the average predicted risk was 9.5% for arrests that occurred after 8 PICU days.

CONCLUSIONS

Patients at high risk of cardiac arrest can be identified with routinely available data after 4 h. The cardiac arrest may occur relatively close to the risk assessment period or days later.

Validation of deep-learning-based triage and acuity score using a large national dataset

AIM

Triage is important in identifying high-risk patients amongst many less urgent patients as emergency department (ED) overcrowding has become a national crisis recently. This study aims to validate that a Deep-learning-based Triage and Acuity Score (DTAS) identifies high-risk patients more accurately than existing triage and acuity scores using a large national dataset.

METHODS

We conducted a retrospective observational cohort study using data from the Korean National Emergency Department Information System (NEDIS), which collected data on visits in real time from 151 EDs. The NEDIS data was split into derivation data (January 2014-June 2016) and validation data (July-December 2016). We also used data from the Sejong General Hospital (SGH) for external validation (January-December 2017). We predicted in-hospital mortality, critical care, and hospitalization using initial information of ED patients (age, sex, chief complaint, time from symptom onset to ED visit, arrival mode, trauma, initial vital signs and mental status as predictor variables).

RESULTS

A total of 11,656,559 patients were included in this study. The primary outcome was in-hospital mortality. The Area Under the Receiver Operating Characteristic curve (AUROC) and Area Under the Precision and Recall Curve (AUPRC) of DTAS were 0.935 and 0.264. It significantly outperformed Korean triage and acuity score (AUROC:0.785, AUPRC:0.192), modified early warning score (AUROC:0.810, AUPRC:0.116), logistic regression (AUROC:0.903, AUPRC:0.209), and random forest (AUROC:0.910, AUPRC:0.179).

CONCLUSION

Deep-learning-based Triage and Acuity Score predicted in-hospital mortality, critical care, and hospitalization more accurately than existing triages and acuity, and it was validated using a large, multicenter dataset.

Practice of diagnosis and management of acute respiratory distress syndrome in mainland China: a cross-sectional study

Background

Although acute respiratory distress syndrome (ARDS) has been recognized for more than 50 years, limited information exists about the incidence and management of ARDS in mainland China. To evaluate the potential for improvement in management of patients with ARDS, this study was designed to describe the incidence and management of ARDS in mainland China.

Methods

, 2012) of all patients who fulfilled the Berlin or American European Consensus Conference (AECC) definition of ARDS in 20 intensive care units, with data collection related to the management of ARDS, patient characteristics and outcomes.

Results

was 90%. A recruitment maneuver was performed in 35.5% of the patients, and 8.7% of patients with severe ARDS received prone position. Overall hospital mortality was 34.0%. Hospital mortality was 21.8% for mild, 31.1% for moderate, and 60.0% for patients with severe ARDS (P=0.004).

Conclusions

Despite general acceptance of low Vt and limited Pplat, high driving pressure, low PEEP and low use of adjunctive measures may still be a concern in mainland China, especially in patients with severe ARDS.

Trial Registration

2012.