On classifying sepsis heterogeneity in the ICU: insight using machine learning

Abstract Objectives Current machine learning models aiming to predict sepsis from electronic health records (EHR) do not account 20 for the heterogeneity of the condition despite its emerging importance in prognosis and treatment. This work demonstrates the added value of stratifying the types of organ dysfunction observed in patients who develop sepsis in the intensive care unit (ICU) in improving the ability to recognize patients at risk of sepsis from their EHR data. Materials and Methods Using an ICU dataset of 13 728 records, we identify clinically significant sepsis subpopulations with distinct organ dysfunction patterns. We perform classification experiments with random forest, gradient boost trees, and support vector machines, using the identified subpopulations to distinguish patients who develop sepsis in the ICU from those who do not. Results The classification results show that features selected using sepsis subpopulations as background knowledge yield a superior performance in distinguishing septic from non-septic patients regardless of the classification model used. The improved performance is especially pronounced in specificity, which is a current bottleneck in sepsis prediction machine learning models. Conclusion Our findings can steer machine learning efforts toward more personalized models for complex conditions including sepsis.

[1]  Teuvo Kohonen,et al.  The self-organizing map , 1990 .

[2]  Jarkko Venna,et al.  Analysis and visualization of gene expression data using Self-Organizing Maps , 2002, Neural Networks.

[3]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[4]  T Laurell,et al.  Classification of motor commands using a modified self-organising feature map. , 2005, Medical engineering & physics.

[5]  Achim Zeileis,et al.  Bias in random forest variable importance measures: Illustrations, sources and a solution , 2007, BMC Bioinformatics.

[6]  Achim Zeileis,et al.  BMC Bioinformatics BioMed Central Methodology article Conditional variable importance for random forests , 2008 .

[7]  C. Sprung,et al.  Surviving Sepsis Campaign: International Guidelines for Management of Severe Sepsis and Septic Shock, 2012 , 2013, Intensive Care Medicine.

[8]  Hana Hazgui,et al.  Ten-year follow-up of cluster-based asthma phenotypes in adults. A pooled analysis of three cohorts. , 2013, American journal of respiratory and critical care medicine.

[9]  Martijn A Spruit,et al.  Clusters of comorbidities based on validated objective measurements and systemic inflammation in patients with chronic obstructive pulmonary disease. , 2013, American journal of respiratory and critical care medicine.

[10]  C. Sprung,et al.  Surviving Sepsis Campaign: International Guidelines for Management of Severe Sepsis and Septic Shock 2012 , 2013, Critical care medicine.

[11]  B. Carr,et al.  Benchmarking the Incidence and Mortality of Severe Sepsis in the United States* , 2013, Critical care medicine.

[12]  Jörn Lötsch,et al.  A machine-learned knowledge discovery method for associating complex phenotypes with complex genotypes. Application to pain , 2013, J. Biomed. Informatics.

[13]  Jing Tian,et al.  Anomaly Detection Using Self-Organizing Maps-Based K-Nearest Neighbor Algorithm , 2014 .

[14]  Malika Charrad,et al.  NbClust: An R Package for Determining the Relevant Number of Clusters in a Data Set , 2014 .

[15]  M. Levy,et al.  Empiric Antibiotic Treatment Reduces Mortality in Severe Sepsis and Septic Shock From the First Hour: Results From a Guideline-Based Performance Improvement Program* , 2014, Critical care medicine.

[16]  Theodore J Iwashyna,et al.  Identifying Patients With Severe Sepsis Using Administrative Claims: Patient-Level Validation of the Angus Implementation of the International Consensus Conference Definition of Severe Sepsis , 2014, Medical care.

[17]  Simon C. Brewer,et al.  Phenotypic clusters within sepsis-associated multiple organ dysfunction syndrome , 2015, Intensive Care Medicine.

[18]  Cheng Soon Ong,et al.  Multivariate spearman's ρ for aggregating ranks using copulas , 2016 .

[19]  R. Bellomo,et al.  The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). , 2016, JAMA.

[20]  J. Vincent,et al.  The Clinical Challenge of Sepsis Identification and Monitoring , 2016, PLoS medicine.

[21]  Peter Szolovits,et al.  MIMIC-III, a freely accessible critical care database , 2016, Scientific Data.

[22]  D. Angus,et al.  Assessment of Global Incidence and Mortality of Hospital-treated Sepsis. Current Estimates and Limitations. , 2016, American journal of respiratory and critical care medicine.

[23]  Bernd Bischl,et al.  mlr: Machine Learning in R , 2016, J. Mach. Learn. Res..

[24]  Uli K. Chettipally,et al.  Prediction of Sepsis in the Intensive Care Unit With Minimal Electronic Health Record Data: A Machine Learning Approach , 2016, JMIR medical informatics.

[25]  Hye Jin Kam,et al.  Learning representations for the early detection of sepsis with deep neural networks , 2017, Comput. Biol. Medicine.

[26]  Stephen L. Jones,et al.  Sepsis as 2 problems: Identifying sepsis at admission and predicting onset in the hospital using an electronic medical record–based acuity score☆,☆☆ , 2017, Journal of critical care.

[27]  Roger G. Mark,et al.  Reproducibility in critical care: a mortality prediction case study , 2017, MLHC.

[28]  Evelyn M. Olenick,et al.  Predicting Sepsis Risk Using the “Sniffer” Algorithm in the Electronic Medical Record , 2017, Journal of nursing care quality.

[29]  S. Lemeshow,et al.  Time to Treatment and Mortality during Mandated Emergency Care for Sepsis , 2017, The New England journal of medicine.

[30]  Ritankar Das,et al.  Reducing patient mortality, length of stay and readmissions through machine learning-based sepsis prediction in the emergency department, intensive care unit and hospital floor units , 2017, BMJ open quality.

[31]  Matthew D. Stanley,et al.  Early sepsis detection in critical care patients using multiscale blood pressure and heart rate dynamics. , 2017, Journal of electrocardiology.

[32]  Florentin Moser,et al.  Poor performance of quick-SOFA (qSOFA) score in predicting severe sepsis and mortality – a prospective study of patients admitted with infection to the emergency department , 2017, Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine.

[33]  Matthew M. Churpek,et al.  Investigating the Impact of Different Suspicion of Infection Criteria on the Accuracy of Quick Sepsis-Related Organ Failure Assessment, Systemic Inflammatory Response Syndrome, and Early Warning Scores* , 2017, Critical care medicine.

[34]  Uli K. Chettipally,et al.  Multicentre validation of a sepsis prediction algorithm using only vital sign data in the emergency department, general ward and ICU , 2018, BMJ Open.

[35]  The Lancet Respiratory Medicine.  Crying wolf: the growing fatigue around sepsis alerts. , 2018, The Lancet. Respiratory medicine.

[36]  Muge Capan,et al.  Data-driven approach to Early Warning Score-based alert management , 2018, BMJ open quality.

[37]  Ron Wehrens,et al.  Flexible Self-Organizing Maps in kohonen 3.0 , 2018 .

[38]  S Maitra,et al.  Accuracy of quick Sequential Organ Failure Assessment (qSOFA) score and systemic inflammatory response syndrome (SIRS) criteria for predicting mortality in hospitalized patients with suspected infection: a meta-analysis of observational studies. , 2018, Clinical microbiology and infection : the official publication of the European Society of Clinical Microbiology and Infectious Diseases.

[39]  Jeremy C. Weiss,et al.  Derivation, Validation, and Potential Treatment Implications of Novel Clinical Phenotypes for Sepsis. , 2019, JAMA.

[40]  C. Coopersmith.  Time to Treatment and Mortality during Mandated Emergency Care for Sepsis , 2019 .

[41]  Chieh-Chen Wu,et al.  Prediction of sepsis patients using machine learning approach: A meta-analysis , 2019, Comput. Methods Programs Biomed..

[42]  Paul Heidenreich,et al.  Electronic health record-based clinical decision support alert for severe sepsis: a randomised evaluation , 2019, BMJ Quality & Safety.