Subject selection
Subjects were eligible for inclusion in the study if they received mechanical ventilation for > 24 hours in the pediatric intensive care unit (PICU), age was less than 18 years, continuous mechanical ventilation data were recorded during that time period and they exhibited hypoxic respiratory failure defined as an oxygen saturation index ³5.(11)
Data collection
Mechanical ventilation was applied using the Servo-I (Getinge AB-Maquet, Gothenburg, Sweden) and connected to a physiologic monitor (IntelliVue MP90, Philips Healthcare, Andover, MA). The mechanical ventilator was interfaced to the monitor using the IntelliBridge medical device-interfacing module (model EC10, Philips Healthcare, Andover). Data was recorded at a frequency of 0.2Hz for the duration of stay in the ICU. The variables included were peak inspiratory pressure (Ppeak), positive end expiratory pressure (PEEP), total respiratory rate (RR), respiratory system compliance (CRS), spontaneous respiratory rate (RRspontaneous), fraction of inspired oxygen (FiO2), expired minute ventilation (Ve), inspired minute ventilation (Vi), spontaneous minute ventilation (Vespontaneous), mean airway pressure (Pmean), end-tidal CO2 concentration (PetCO2), volumetric CO2 elimination (VCO2), expired tidal volume (Vte), inspired tidal volume (Vti), estimate of the pressure in the first 100ms of the breath (P100), end-expiratory flow rate (Vee), work of breathing of the ventilator (WOBvent), barometric pressure (Pb), heart rate (HR), oxygen saturation (SpO2) and dead-space fraction (Vd/Vt).
Demographic and outcome data were abstracted from the medical record for each subject and the diagnosis was recorded according to the International Classification of Diseases published by the World Health Organization (Revisions 9 and 10, Clinical Modification) and binned to either primary respiratory, surgical procedure, neurologic, sepsis or other.(14)
Data preprocessing
Because the procedures included in computable phenotype extraction often require the input data to be completely intact (no gaps in data), 1-dimensional linear interpolation was implemented for each variable. Further, the physiologic monitor and mechanical ventilator offer built-in preprocessing but signals can still be corrupted by noise and artifact.(15) Band-pass and low-pass filters were applied to filter out data that was not physiologically plausible according to established methods.(16) A Savitzky-Golay filter was applied to mechanical ventilation data in order to remove noise and artifact but preserve local data phenomenon.(17) Further, normalization of individual parameters is important in a pediatric population since signals are expected to change as the child grows. Data were normalized to either body weight (for respiratory parameters tidal volume, minute ventilation, carbon dioxide elimination, end-expiratory flow rate) and Z-scores were computed for heart rate and respiratory rate.(16, 18)
Case identification
For an individual subject, a case was defined as a 2-hour period, to include a 1-hour period preceding and 1-hour period following a change in PEEP. We have previously demonstrated that a period ~60 minutes is necessary to observe physiologic effects from modest changes in PEEP.(19) A quality function was built to ensure that only ‘clean’ cases were analyzed. A clean PEEP case was defined as one where no ventilator changes were made (other than PEEP and FiO2); the PEEP change was sustained for > 1-hour. For cases where the PEEP was increased, a responder was defined as a case that exhibited any improvement in Vd/Vt; [Vd/Vtpost – Vd/Vpre] > 0. For cases where PEEP was decreased, a responder was defined as a case where oxygenation was maintained; [Vd/Vtpost – Vd/Vpre] ³ 0.
Computable phenotype extraction
In most clinical investigations involving mechanical ventilation data, descriptive statistics are typically computed. However, since time-series data recorded at a relatively high frequency (compared to data transcribed in the electronic medical record), various representations of individual signals can be computed. Although thousands of methods exist in the literature, a set of 219 methods (including autocorrelation, auto-mutual information, stationarity, entropy, correlation dimension, linear and non-linear model fits, measures from the power spectrum and outlier quantification) provides a robust summary of the different behaviors of time series analysis.(20)
HCTSA has recently been applied in some areas of bioengineering, but applications on medical time-series data have not yet been studied.(21) We quantify a wide range of time series properties, computable phenotypes according to existing methods.(20)
Feature selection and model training
A small number of computable phenotypes were selected from the total set of phenotypes by implementing the least absolute shrinkage and selection operator according to established methods (LASSO).(22) A support vector machine (SVM) is a supervised machine learning algorithm that is applied to a wide variety of pattern recognition and classification problems and achieves good discriminative power in various healthcare data applications.(23-25) A separate model was trained for all cases where the PEEP was increased (PEEPSVM) and where the PEEP was decreased (PEEP¯SVM) since it there are important differences in the disease acuity at times when PEEP would be increased or decreased. To protect against overfitting and to assess model performance a 5-fold cross validation was performed.
Statistical analyses
The D’Agostino and Pearson omnibus test was applied to test the normality of the data. Since the data were not normally distributed, continuous variables are presented as median (interquartile range). To assess model performance, the area under the receiver operator characteristic curve was calculated as well as the diagnostic accuracy, sensitivity, specificity, positive and negative predictive values and the positive and negative likelihood ratios. Data aggregation, cleaning and analyses were conducted using MATLAB (V9.1.0.441655, The Mathworks Inc., Natick, MA).
The protocol was approved by the Boston Children’s Hospital Institutional Review Board and need for informed consent was waived.