Predicting long-term mortality with first week post-operative data after Coronary Artery Bypass Grafting using Machine Learning models

Coronary Artery Bypass Graft (CABG) surgery is the most common cardiac operation and its complications are associated with increased long-term mortality rates. Although many factors are known to be linked to this, much remains to be understood about their exact influence on outcome. In this study we used Machine Learning (ML) algorithms to predict long-term mortality in CABG patients using data from routinely measured clinical parameters from a large cohort of CABG patients (n=5868). We compared the accuracy of 5 different ML models with traditional Cox and Logistic Regression, and report the most important variables in the best performing models. In the validation dataset, the Gradient Boosted Machine (GBM) algorithm was the most accurate (AUROC curve [95%CI] of 0.767 [0.739-0.796]), proving to be superior to traditional Cox and logistic regression (p <0.01) for long-term mortality prediction. Measures of variable importance for outcome prediction extracted from the GBM and Random Forest models partly reflected what is known in the literature, but interestingly also highlighted other unexpectedly relevant parameters. In conclusion, we found ML algorithm-based models to be more accurate than traditional Logistic Regression in predicting long-term mortality after CABG. Finally, these models may provide essential input to assist the development of intelligent decision support systems for clinical use.

[1]  Eva K. Lee,et al.  Rationale and methodology of a collaborative learning project in congenital cardiac care. , 2016, American heart journal.

[2]  Vasile Palade,et al.  Class Imbalance Learning Methods for Support Vector Machines , 2013 .

[3]  Alexander Zien,et al.  Semi-Supervised Classification by Low Density Separation , 2005, AISTATS.

[4]  Ramón Díaz-Uriarte,et al.  Gene selection and classification of microarray data using random forest , 2006, BMC Bioinformatics.

[5]  Cyril Ferdynus,et al.  A Comparison of a Machine Learning Model with EuroSCORE II in Predicting Mortality after Elective Cardiac Surgery: A Decision Curve Analysis , 2017, PloS one.

[6]  William Fleischman,et al.  Prediction of In-hospital Mortality in Emergency Department Patients With Sepsis: A Local Big Data-Driven, Machine Learning Approach. , 2016, Academic emergency medicine : official journal of the Society for Academic Emergency Medicine.

[7]  Thomas Berg,et al.  FibroGENE: A gene-based model for staging liver fibrosis. , 2016, Journal of hepatology.

[8]  Hendriek C Boshuizen,et al.  Toward Targeted Hypertension Screening Guidelines , 2006, Medical decision making : an international journal of the Society for Medical Decision Making.

[9]  Chao Chen,et al.  Using Random Forest to Learn Imbalanced Data , 2004 .

[10]  Trevor Hastie,et al.  The Elements of Statistical Learning , 2001 .

[11]  Stephen Kwek,et al.  Applying Support Vector Machines to Imbalanced Datasets , 2004, ECML.

[12]  Chih-Jen Lin,et al.  A Practical Guide to Support Vector Classication , 2008 .

[13]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[14]  Greg Ridgeway,et al.  Generalized Boosted Models: A guide to the gbm package , 2006 .

[15]  Andy Liaw,et al.  Classification and Regression by randomForest , 2007 .

[16]  Sean M. O'Brien,et al.  Predictors of Long-Term Survival After Coronary Artery Bypass Grafting Surgery: Results From the Society of Thoracic Surgeons Adult Cardiac Surgery Database (The ASCERT Study) , 2012, Circulation.

[17]  Anne-Laure Boulesteix,et al.  Overview of random forest methodology and practical guidance with emphasis on computational biology and bioinformatics , 2012, WIREs Data Mining Knowl. Discov..

[18]  Mohammad Khalilia,et al.  Predicting disease risks from highly imbalanced data using random forest , 2011, BMC Medical Informatics Decis. Mak..

[19]  J. Carpenter,et al.  Practice of Epidemiology Comparison of Random Forest and Parametric Imputation Models for Imputing Missing Data Using MICE: A CALIBER Study , 2014 .

[20]  Gerasimos Filippatos,et al.  Prognostic value of blood urea nitrogen in patients hospitalized with worsening heart failure: insights from the Acute and Chronic Therapeutic Impact of a Vasopressin Antagonist in Chronic Heart Failure (ACTIV in CHF) study. , 2007, Journal of cardiac failure.

[21]  J. Rouleau,et al.  Sustained postoperative anaemia is associated with an impaired outcome after coronary artery bypass graft surgery: insights from the IMAGINE trial , 2011, Heart.

[22]  Jeremy Miles,et al.  Discovering statistics using R, 1st Edition , 2012 .

[23]  J. Friedman.  Greedy function approximation: A gradient boosting machine. , 2001 .

[24]  Christophe Mues,et al.  An experimental comparison of classification algorithms for imbalanced credit scoring data sets , 2012, Expert Syst. Appl..

[25]  Max Kuhn,et al.  caret: Classification and Regression Training , 2015 .

[26]  Ulrike von Luxburg,et al.  Optimal construction of k-nearest-neighbor graphs for identifying noisy clusters , 2009, Theoretical Computer Science.

[27]  J. Schafer,et al.  Missing data: our view of the state of the art. , 2002, Psychological methods.

[28]  Gérard Biau,et al.  Analysis of a Random Forests Model , 2010, J. Mach. Learn. Res..

[29]  K. Borgwardt,et al.  Machine Learning in Medicine , 2015, Mach. Learn. under Resour. Constraints Vol. 3.

[30]  R. Henning,et al.  Immediate postoperative renal function deterioration in cardiac surgical patients predicts in-hospital mortality and long-term survival. , 2004, Journal of the American Society of Nephrology : JASN.

[31]  Constantine Frangakis,et al.  Multiple imputation by chained equations: what is it and how does it work? , 2011, International journal of methods in psychiatric research.

[32]  Daniel Polsky,et al.  Coronary revascularization trends in the United States, 2001-2008. , 2011, JAMA.

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

[34]  Xavier Robin,et al.  pROC: an open-source package for R and S+ to analyze and compare ROC curves , 2011, BMC Bioinformatics.

[35]  Klaus Hechenbichler,et al.  Weighted k-Nearest-Neighbor Techniques and Ordinal Classification , 2004 .

[36]  E. DeLong,et al.  Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. , 1988, Biometrics.

[37]  A. Epema,et al.  Postoperative renal dysfunction and preoperative left ventricular dysfunction predispose patients to increased long-term mortality after coronary artery bypass graft surgery. , 2009, British journal of anaesthesia.

[38]  Nan Xiao,et al.  hdnom: Building Nomograms for Penalized Cox Models with High-Dimensional Survival Data , 2016, bioRxiv.

[39]  Romain Pirracchio,et al.  Incidence, risk factors and prediction of post-operative acute kidney injury following cardiac surgery for active infective endocarditis: an observational study , 2013, Critical Care.

[40]  L. Mbuagbaw,et al.  A tutorial on sensitivity analyses in clinical trials: the what, why, when and how , 2013, BMC Medical Research Methodology.

[41]  David O. Meltzer,et al.  Multicenter Comparison of Machine Learning Methods and Conventional Regression for Predicting Clinical Deterioration on the Wards , 2016, Critical care medicine.

[42]  William T. Abraham,et al.  Risk stratification for in-hospital mortality in acutely decompensated heart failure. Classification and regression tree analysis , 2005 .

[43]  Stef van Buuren,et al.  MICE: Multivariate Imputation by Chained Equations in R , 2011 .

[44]  Alois Knoll,et al.  Gradient boosting machines, a tutorial , 2013, Front. Neurorobot..