Dynamic Autoregressive Modelling of Critical Care Patients as a Basis for Health Monitoring

Real-time modelling techniques could be valuable to continuously evaluate individual critically ill patients and to help the medical staff with estimation of prognosis. This preliminary study examines the possibilities to distinguish survivors from non-survivors on the basis of instabilities in the dynamics of daily measured variables. A data set, containing 140 patients, was generated in the intensive care unit (ICU) of the university hospital of Leuven. First and second order dynamic auto-regression (DAR) models were used to quantify the stability of time series of three physiological variables as a criterion to distinguish survivors from non-survivors. The best results were found for blood urea concentration with true negative fractions of 45/72 (63%) and true positive fractions of 43/68 (62%). The results indicate that the dynamics of time series of laboratory parameters from critically ill patients are indicative for their clinical condition and outcome.

[1]  N. Seleno,et al.  Elevation of Blood Urea Nitrogen is Predictive of Long-term Mortality in Critically Ill Patients Independent of “Normal” Creatinine , 2011 .

[2]  Ursula Gather,et al.  Time-Effect Relations of Medical Interventions in a Clinical Information System , 1999, KI.

[3]  R. G. Spencer,et al.  Self-organising discovery, recognition and prediction of haemodynamic patterns in the intensive care unit , 1997, Medical and Biological Engineering and Computing.

[4]  L. Lipsitz.  Dynamics of stability: the physiologic basis of functional health and frailty. , 2002, The journals of gerontology. Series A, Biological sciences and medical sciences.

[5]  Diego J. Pedregal,et al.  Environmental time series analysis and forecasting with the Captain toolbox , 2007, Environ. Model. Softw..

[6]  Tobias Keck,et al.  Elevated blood urea nitrogen is an independent risk factor of prolonged intensive care unit stay due to acute necrotizing pancreatitis. , 2010, Journal of critical care.

[7]  Peter C. Young,et al.  System identification, time series analysis and forecasting : The Captain Toolbox handbook. , 2007 .

[8]  Timothy G Buchman,et al.  Nonlinear dynamics, complex systems, and the pathobiology of critical illness , 2004, Current opinion in critical care.

[9]  D C McMillan,et al.  Postoperative serum urea is associated with 30-day mortality in patients undergoing emergency abdominal surgery , 2006, Annals of clinical biochemistry.

[10]  R. Chang,et al.  Predicting deaths among intensive care unit patients. , 1988, Critical care medicine.

[11]  Leon Glass,et al.  Dynamical disease: Identification, temporal aspects and treatment strategies of human illness. , 1995, Chaos.

[12]  Jean-Marie Aerts,et al.  Prediction of Clinical Conditions after Coronary Bypass Surgery using Dynamic Data Analysis , 2010, Journal of Medical Systems.

[13]  Gilles Clermont,et al.  Dynamic microsimulation to model multiple outcomes in cohorts of critically ill patients , 2004, Intensive Care Medicine.

[14]  P. Young,et al.  Time series analysis, forecasting and control , 1972, IEEE Transactions on Automatic Control.

[15]  Stuart M. Cobbe,et al.  375 Does blood urea concentration predict early mortality in patients hospitalised with decompensated Heart Failure (HF) better than estimated glomerular filtration rate (eGFR) , 2008 .

[16]  James McNames,et al.  Physiologic data acquisition system and database for the study of disease dynamics in the intensive care unit* , 2003, Critical care medicine.

[17]  R. Chang,et al.  Predicting outcome among intensive care unit patients using computerised trend analysis of daily Apache II scores corrected for organ system failure , 1988, Intensive Care Medicine.

[18]  Cristina Ribeiro,et al.  Vital Signs in Intensive Care: Automatic Acquisition and Consolidation into Electronic Patient Records , 2009, Journal of Medical Systems.

[19]  Tudor Toma,et al.  Discovery and inclusion of SOFA score episodes in mortality prediction , 2007, J. Biomed. Informatics.

[20]  Scott K Aberegg,et al.  Intensive insulin therapy in the medical ICU. , 2006, The New England journal of medicine.

[21]  L. Glass.  Synchronization and rhythmic processes in physiology , 2001, Nature.

[22]  C. Lambert,et al.  Time-series analysis of long-term ambulatory myocardial ischemia: effects of beta-adrenergic and calcium channel blockade. , 1995, American heart journal.