Multiscale network representation of physiological time series for early prediction of sepsis

Abstract

Sepsis, a dysregulated immune-mediated host response to infection, is the leading cause of morbidity and mortality in critically ill patients. Indices of heart rate variability and complexity (such as entropy) have been proposed as surrogate markers of neuro-immune system dysregulation with diseases such as sepsis. However, these indices only provide an average, one dimensional description of complex neuro-physiological interactions. We propose a novel multiscale network construction and analysis method for multivariate physiological time series, and demonstrate its utility for early prediction of sepsis. We show that features derived from a multiscale heart rate and blood pressure time series network provide approximately 20% improvement in the area under the receiver operating characteristic (AUROC) for four hours ahead prediction of sepsis over traditional indices of heart rate entropy (0.78 versus 0.66). Our results indicate that this improvement is attributable to both the improved network construction method proposed here, as well as the information embedded in the higher order interaction of heart rate and blood pressure time series dynamics. Our final model, which included the most commonly available clinical measurements in patients’ Electronic Medical Records, multiscale entropy features as well as the proposed network-based features, achieved an AUROC of 0.80. Prediction of the onset of sepsis prior to clinical recognition will allow for meaningful earlier interventions (e.g., antibiotic and fluid administration), which have the potential to decrease sepsis-related morbidity, mortality and healthcare costs.

Machine learning models for early sepsis recognition in the neonatal intensive care unit using readily available electronic health record data

“…However, in a subsequent, large retrospective study, the HeRO score failed to detect neonatal sepsis, suggesting the predictive value is uncertain in clinical practice [33]. Additional studies used a novel network representation of vital sign dynamics [22,23] for sepsis prediction in adults. However, these models require input features derived from heart rate measurements collected every few seconds from bedside monitors, which are not typically available in most EHRs thereby limiting their general applicability.…”

The effect of fractal-like mechanical ventilation on vital signs in a rat model of acute-on-chronic liver failure

“…Development of new analytical methods in Network Physiology will also help to uncover the mechanism of multiple organ failure in cirrhosis form a different perspective (Bartsch et al 2015, Xiong et al 2017, Asada et al 2016, Ivanov et al 2016, Kanter et al 2015. For example, multiscale network construction (Shashikumar et al 2017) can be employed for quantification of cardiovascular and respiratory interaction within the context of sepsis in patients with liver failure. Such novel approaches are crucial as classical medicine currently lacks a good pathophysiological model to explain the mechanism of multiple organ failure in acute-on-chronic liver failure.…”

Machine learning of physiological waveforms and electronic health record data to predict, diagnose and treat haemodynamic instability in surgical patients: protocol for a retrospective study

“…Some of these approaches have been used in the ICU setting to predict sepsis but have never been used in the perioperative/surgical setting 8–11. We have successfully applied such methods to problems involving large amounts of multivariate time series data 20.…”

“…As a result, the potential that hospital data offer in terms of understanding and improving care has not been realised. While physiological prediction tools have been developed in the critical care setting,8–11 the goal of this proposal is to develop, validate and test real-time intraoperative risk prediction tools based on EHR data and high-fidelity physiological waveforms to predict cardiorespiratory instability (CRI) in the perioperative/surgical setting. New onset of CRI is common in patients undergoing surgery.…”