Gleaning knowledge from data in the intensive care unit.

It is often difficult to accurately predict when, why, and which patients develop shock, because signs of shock often occur late, once organ injury is already present. Three levels of aggregation of information can be used to aid the bedside clinician in this task: analysis of derived parameters of existing measured physiologic variables using simple bedside calculations (functional hemodynamic monitoring); prior physiologic data of similar subjects during periods of stability and disease to define quantitative metrics of level of severity; and libraries of responses across large and comprehensive collections of records of diverse subjects whose diagnosis, therapies, and course is already known to predict not only disease severity, but also the subsequent behavior of the subject if left untreated or treated with one of the many therapeutic options. The problem is in defining the minimal monitoring data set needed to initially identify those patients across all possible processes, and then specifically monitor their responses to targeted therapies known to improve outcome. To address these issues, multivariable models using machine learning data-driven classification techniques can be used to parsimoniously predict cardiorespiratory insufficiency. We briefly describe how these machine learning approaches are presently applied to address earlier identification of cardiorespiratory insufficiency and direct focused, patient-specific management.

[1]  W. Knaus,et al.  The APACHE III prognostic system. Risk prediction of hospital mortality for critically ill hospitalized adults. , 1991, Chest.

[2]  H. Montenegro,et al.  Beyond the intensive care unit: a review of interventions aimed at anticipating and preventing in-hospital cardiopulmonary arrest. , 2007, Resuscitation.

[3]  Sven Zenker,et al.  Use of non-invasive NIRS during a vascular occlusion test to assess dynamic tissue O2 saturation response , 2008, Intensive Care Medicine.

[4]  Stanley Lemeshow,et al.  The Logistic Organ Dysfunction System , 1997 .

[5]  J. Vincent,et al.  The prognostic value of muscle StO2 in septic patients , 2007, Intensive Care Medicine.

[6]  Roman Garnett,et al.  Mining intensive care vitals for leading indicators of adverse health events , 2011 .

[7]  C. Subbe,et al.  Validation of a modified Early Warning Score in medical admissions. , 2001, QJM : monthly journal of the Association of Physicians.

[8]  H. Smithline,et al.  A comparison of the shock index and conventional vital signs to identify acute, critical illness in the emergency department. , 1994, Annals of emergency medicine.

[9]  Marilyn Hravnak,et al.  Cardiorespiratory instability before and after implementing an integrated monitoring system* , 2011, Critical care medicine.

[10]  S Lemeshow,et al.  The Logistic Organ Dysfunction system. A new way to assess organ dysfunction in the intensive care unit. ICU Scoring Group. , 1996, JAMA.

[11]  M. Pinsky Complexity modeling: Identify instability early , 2010, Critical care medicine.

[12]  D. Koller,et al.  Integration of Early Physiological Responses Predicts Later Illness Severity in Preterm Infants , 2010, Science Translational Medicine.

[13]  M. Pinsky,et al.  Thenar oxygen saturation during weaning from mechanical ventilation: an observational study , 2013, European Respiratory Journal.

[14]  J. Vincent,et al.  Circulatory shock. , 2013, The New England journal of medicine.

[15]  David A. Clifton,et al.  Gaussian process clustering for the functional characterisation of vital-sign trajectories , 2013, 2013 IEEE International Workshop on Machine Learning for Signal Processing (MLSP).

[16]  Didier Payen,et al.  Functional hemodynamic monitoring. , 2002, Intensive care medicine.

[17]  J. Vincent,et al.  The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure , 1996, Intensive Care Medicine.

[18]  D. Chemla,et al.  Relation between respiratory changes in arterial pulse pressure and fluid responsiveness in septic patients with acute circulatory failure. , 2000, American journal of respiratory and critical care medicine.

[19]  M. Pinsky,et al.  Prognostic implications of tissue oxygen saturation in human septic shock , 2012, Intensive Care Medicine.

[20]  S. Magder Central venous pressure: A useful but not so simple measurement , 2006, Critical care medicine.

[21]  W. Benitz Mortality Reduction by Heart Rate Characteristic Monitoring in Very Low Birth Weight Neonates: A Randomized Trial , 2012 .

[22]  A. G. Cano,et al.  Dynamic arterial elastance to predict arterial pressure response to volume loading in preload-dependent patients , 2011 .

[23]  F. Ismail,et al.  Integrated monitoring and analysis for early warning of patient deterioration. , 2007, British journal of anaesthesia.