Associations Between End-Tidal Carbon Dioxide During Pediatric Cardiopulmonary Resuscitation, Cardiopulmonary Resuscitation Quality, and Survival.

BACKGROUND: Supported by laboratory and clinical investigations of adult cardiopulmonary arrest, resuscitation guidelines recommend monitoring end-tidal carbon dioxide (ETCO2) as an indicator of cardiopulmonary resuscitation (CPR) quality, but they note that "specific values to guide therapy have not been established in children." METHODS: This prospective observational cohort study was a National Heart, Lung, and Blood Institute-funded ancillary study of children in the ICU-RESUS trial (Intensive Care Unit-Resuscitation Project; NCT02837497). Hospitalized children (≤18 years of age and ≥37 weeks postgestational age) who received chest compressions of any duration for cardiopulmonary arrest, had an endotracheal or tracheostomy tube at the start of CPR, and evaluable intra-arrest ETCO2 data were included. The primary exposure was event-level average ETCO2 during the first 10 minutes of CPR (dichotomized as ≥20 mm Hg versus <20 mm Hg on the basis of adult literature). The primary outcome was survival to hospital discharge. Secondary outcomes were sustained return of spontaneous circulation, survival to discharge with favorable neurological outcome, and new morbidity among survivors. Poisson regression measured associations between ETCO2 and outcomes as well as the association between ETCO2 and other CPR characteristics: (1) invasively measured systolic and diastolic blood pressures, and (2) CPR quality and chest compression mechanics metrics (ie, time to CPR start; chest compression rate, depth, and fraction; ventilation rate). RESULTS: Among 234 included patients, 133 (57%) had an event-level average ETCO2 ≥20 mm Hg. After controlling for a priori covariates, average ETCO2 ≥20 mm Hg was associated with a higher incidence of survival to hospital discharge (86/133 [65%] versus 48/101 [48%]; adjusted relative risk, 1.33 [95% CI, 1.04-1.69]; P=0.023) and return of spontaneous circulation (95/133 [71%] versus 59/101 [58%]; adjusted relative risk, 1.22 [95% CI, 1.00-1.49]; P=0.046) compared with lower values. ETCO2 ≥20 mm Hg was not associated with survival with favorable neurological outcome or new morbidity among survivors. Average 2 ≥20 mm Hg was associated with higher systolic and diastolic blood pressures during CPR, lower CPR ventilation rates, and briefer pre-CPR arrest durations compared with lower values. Chest compression rate, depth, and fraction did not differ between ETCO2 groups. CONCLUSIONS: In this multicenter study of children with in-hospital cardiopulmonary arrest, ETCO2 ≥20 mm Hg was associated with better outcomes and higher intra-arrest blood pressures, but not with chest compression quality metrics.

A machine learning approach to predict quality of life changes in patients with Parkinson's Disease.

OBJECTIVE: Parkinson disease (PD) is a progressive neurodegenerative disorder with an annual incidence of approximately 0.1%. While primarily considered a motor disorder, increasing emphasis is being placed on its non-motor features. Both manifestations of the disease affect quality of life (QoL), which is captured in part II of the Unified Parkinson's Disease Rating Scale (UPDRS-II). While useful in the management of patients, it remains challenging to predict how QoL will change over time in PD. The goal of this work is to explore the feasibility of a machine learning algorithm to predict QoL changes in PD patients. METHODS: In this retrospective cohort study, patients with at least 12 months of follow-up were identified from the Parkinson's Progression Markers Initiative database (N = 630) and divided into two groups: those with and without clinically significant worsening in UPDRS-II (n = 404 and n = 226, respectively). We developed an artificial neural network using only UPDRS-II scores, to predict whether a patient would clinically worsen or not at 12 months from follow-up. RESULTS: Using UPDRS-II at baseline, at 2 months, and at 4 months, the algorithm achieved 90% specificity and 56% sensitivity. INTERPRETATION: A learning model has the potential to rule in patients who may exhibit clinically significant worsening in QoL at 12 months. These patients may require further testing and increased focus.

Prediction of Periventricular Leukomalacia in Neonates after Cardiac Surgery Using Machine Learning Algorithms.

Periventricular leukomalacia (PVL) is brain injury that develops commonly in neonates after cardiac surgery. Earlier identification of patients who are at higher risk for PVL may improve clinicians' ability to optimize care for these challenging patients. The aim of this study was to apply machine learning algorithms and wavelet analysis to vital sign and laboratory data obtained from neonates immediately after cardiac surgery to predict PVL occurrence. We analyzed physiological data of patients with and without hypoplastic left heart syndrome (HLHS) during the first 12 h after cardiac surgery. Wavelet transform was applied to extract time-frequency information from the data. We ranked the extracted features to select the most discriminative features, and the support vector machine with radial basis function as a kernel was selected as the classifier. The classifier was optimized via three methods: (1) mutual information, (2) modified mutual information considering the reliability of features, and (3) modified mutual information with reliability index and maximizing set's mutual information. We assessed the accuracy of the classifier at each time point. A total of 71 neonates met the study criteria. The rates of PVL occurrence were 33% for all patients, with 41% in the HLHS group and 25% in the non-HLHS group. The F-score results for HLHS patients and non-HLHS patients were 0.88 and 1.00, respectively. Using maximizing set's mutual information improved the classifier performance in the all patient groups from 0.69 to 0.81. The novel application of a modified mutual information ranking system with the reliability index in a PVL prediction model provided highly accurate identification. This tool is a promising step for improving the care of neonates who are at higher risk for developing PVL following cardiac surgery.

A Dynamic Model of Rescuer Parameters for Optimizing Blood Gas Delivery during Cardiopulmonary Resuscitation.

INTRODUCTION: The quality of cardiopulmonary resuscitation (CPR) has been shown to impact patient outcomes. However, post-CPR morbidity and mortality remain high, and CPR optimization is an area of active research. One approach to optimizing CPR involves establishing reliable CPR performance measures and then modifying CPR parameters, such as compressions and ventilator breaths, to enhance these measures. We aimed to define a reliable CPR performance measure, optimize the CPR performance based on the defined measure and design a dynamically optimized scheme that varies CPR parameters to optimize CPR performance. MATERIALS AND METHODS: We selected total blood gas delivery (systemic oxygen delivery and carbon dioxide delivery to the lungs) as an objective function for maximization. CPR parameters were divided into three categories: rescuer dependent, patient dependent, and constant parameters. Two optimization schemes were developed using simulated annealing method: a global optimization scheme and a sequential optimization scheme. RESULTS AND DISCUSSION: Variations of CPR parameters over CPR sequences (cycles) were analyzed. Across all patient groups, the sequential optimization scheme resulted in significant enhancement in the effectiveness of the CPR procedure when compared to the global optimization scheme. CONCLUSIONS: Our study illustrates the potential benefit of considering dynamic changes in rescuer-dependent parameters during CPR in order to improve performance. The advantage of the sequential optimization technique stemmed from its dynamically adapting effect. Our CPR optimization findings suggest that as CPR progresses, the compression to ventilation ratio should decrease, and the sequential optimization technique can potentially improve CPR performance. Validation in vivo is needed before implementing these changes in actual practice.

Identification of sympathetic and parasympathetic nerves function in cardiovascular regulation using ANFIS approximation.

OBJECTIVE: In this paper a new nonlinear system identification approach is developed for dynamical quantification of cardiovascular regulation. This approach is specifically focused on the identification of the heart rate (HR) baroreflex mechanism. The principal objective of this paper is to improve the model accuracy in the estimation of HR by proposing a modified nonlinear model. METHODS AND MATERIAL: The proposed HR baroreflex model is based on inherent features of the autonomic nervous system for which we develop an adaptive neuro-fuzzy inference system (ANFIS) structure. This method allows incorporation of physiological understandings about the sympathetic and parasympathetic nerves through the selection of appropriate membership functions in the ANFIS structure. The required data for system modeling are collected from the publicly available PhysioNet database. RESULTS: The results agree with the natural characteristics and physiological understanding of the cardiovascular regulatory system, such as delay in the parasympathetic function, durability in the function of sympathetic nerves and the correlation between the HR and the ABP signals. They also show significant improvements in HR prediction in terms of the normalized root mean square error (NRMSE) in comparison with other reported methods. We achieved to 0.191 in mean NRMSE in prediction of HR in this paper which is about 20% better than the best reported result in other researches. CONCLUSION: We have shown that for cardiovascular system regulation, our proposed nonlinear model is more accurate than other recently developed methods. Accurate HR baroreflex modeling enables clinicians to have more reliable information for their patients.

Prediction of periventricular leukomalacia. Part I: Selection of hemodynamic features using logistic regression and decision tree algorithms.

OBJECTIVE: Periventricular leukomalacia (PVL) is part of a spectrum of cerebral white matter injury which is associated with adverse neurodevelopmental outcome in preterm infants. While PVL is common in neonates with cardiac disease, both before and after surgery, it is less common in older infants with cardiac disease. Pre-, intra-, and postoperative risk factors for the occurrence of PVL are poorly understood. The main objective of the present work is to identify potential hemodynamic risk factors for PVL occurrence in neonates with complex heart disease using logistic regression analysis and decision tree algorithms. METHODS: The postoperative hemodynamic and arterial blood gas data (monitoring variables) collected in the cardiac intensive care unit of Children's Hospital of Philadelphia were used for predicting the occurrence of PVL. Three categories of datasets for 103 infants and neonates were used-(1) original data without any preprocessing, (2) partial data keeping the admission, the maximum and the minimum values of the monitoring variables, and (3) extracted dataset of statistical features. The datasets were used as inputs for forward stepwise logistic regression to select the most significant variables as predictors. The selected features were then used as inputs to the decision tree induction algorithm for generating easily interpretable rules for prediction of PVL. RESULTS: Three sets of data were analyzed in SPSS for identifying statistically significant predictors (p<0.05) of PVL through stepwise logistic regression and their correlations. The classification success of the Case 3 dataset of extracted statistical features was best with sensitivity (SN), specificity (SP) and accuracy (AC) of 87, 88 and 87%, respectively. The identified features, when used with decision tree algorithms, gave SN, SP and AC of 90, 97 and 94% in training and 73, 58 and 65% in test. The identified variables in Case 3 dataset mainly included blood pressure, both systolic and diastolic, partial pressures pO(2) and pCO(2), and their statistical features like average, variance, skewness (a measure of asymmetry) and kurtosis (a measure of abrupt changes). Rules for prediction of PVL were generated automatically through the decision tree algorithms. CONCLUSIONS: The proposed approach combines the advantages of statistical approach (regression analysis) and data mining techniques (decision tree) for generation of easily interpretable rules for PVL prediction. The present work extends an earlier research [Galli KK, Zimmerman RA, Jarvik GP, Wernovsky G, Kuijpers M, Clancy RR, et al. Periventricular leukomalacia is common after cardiac surgery. J Thorac Cardiovasc Surg 2004;127:692-704] in the form of expanding the feature set, identifying additional prognostic factors (namely pCO(2)) emphasizing the temporal variations in addition to upper or lower values, and generating decision rules. The Case 3 dataset was further investigated in Part II for feature selection through computational intelligence.

Prediction of periventricular leukomalacia. Part II: Selection of hemodynamic features using computational intelligence.

OBJECTIVE: The objective of Part II is to analyze the dataset of extracted hemodynamic features (Case 3 of Part I) through computational intelligence (CI) techniques for identification of potential prognostic factors for periventricular leukomalacia (PVL) occurrence in neonates with congenital heart disease. METHODS: The extracted features (Case 3 dataset of Part I) were used as inputs to CI based classifiers, namely, multi-layer perceptron (MLP) and probabilistic neural network (PNN) in combination with genetic algorithms (GA) for selection of the most suitable features predicting the occurrence of PVL. The selected features were next used as inputs to a decision tree (DT) algorithm for generating easily interpretable rules of PVL prediction. RESULTS: Prediction performance for two CI based classifiers, MLP and PNN coupled with GA are presented for different number of selected features. The best prediction performances were achieved with 6 and 7 selected features. The prediction success was 100% in training and the best ranges of sensitivity (SN), specificity (SP) and accuracy (AC) in test were 60-73%, 74-84% and 71-74%, respectively. The identified features when used with the DT algorithm gave best SN, SP and AC in the ranges of 87-90% in training and 80-87%, 74-79% and 79-82% in test. Among the variables selected in CI, systolic and diastolic blood pressures, and pCO(2) figured prominently similar to Part I. Decision tree based rules for prediction of PVL occurrence were obtained using the CI selected features. CONCLUSIONS: The proposed approach combines the generalization capability of CI based feature selection approach and generation of easily interpretable classification rules of the decision tree. The combination of CI techniques with DT gave substantially better test prediction performance than using CI and DT separately.