A real-time, continuous physiologic data acquisition system for the study of dynamical disease in the intensive care unit

Abstract

Introduction: The study of dynamical disease, or how disease states change with respect to time, is key to understanding the underlying pathophysiological control mechanisms involved in critical illness. Collection of data as a function of time is the first step in obtaining a dynamical perspective of disease. However, collection of long (hours-days) time series, in the intensive care unit (ICU) environment, has proven difficult. We describe a real-time, data acquisition system for the study of dynamical disease in the ICU that captures physiological data from 66 HP-Merlin digital bedside monitors located in three ICUs and the operating rooms at the University of Pittsburgh and the Pediatric ICU at Oregon Health Sciences University. Methods: The system is based on four components: 1) 66 Hewlett-Packard (HP) Merlin patient monitors located in patient rooms and connected via a SDN network to; 2) HP Patient Data Servers (PDS) connected via LAN using standard TCP-IP protocols to; 3) HP workstations located in each institution's computer laboratory. Both real-time waveform (EGG, EEG, respiration, and pressures [systemic, pulmonary, central venous, intracranial]), and parameter (time-averaged mean values) data are acquired continuously and saved in binary multiple-channel file format for display and subsequent linear and nonlinear time series analyses including mean, standard deviation, power spectrum, fractal analysis, approximate entropy, wavelet analysis, and other nonlinear metrics. The systems continuously capture data on a 24 hour × 7 day schedule. Results: To date, we have successfully recorded real-time continuous time series data from > 30 ICU patients. These data are compressed and stored in physiologic databases of specific disease states (e.g. brain injury, sepsis/septic shock). Conclusions: This computerized system is the first step in real-time study of critical illness and injury from a dynamic perspective. Future development of the system will include real time and post hoc computations for determination of linear and nonlinear metrics used to characterize, evaluate, and predict specific ICU disease states.