Machine Learning Methods for Predicting Long-Term Mortality in Patients After Cardiac Surgery

Abstract

Objective:This study aims to construct and validate several machine learning (ML) algorithms to predict long-term mortality and identify risk factors in unselected patients post-cardiac surgery.MethodsThe Medical Information Mart for Intensive Care (MIMIC-III) database was used to perform a retrospective administrative database study. Candidate predictors consisted of the demographics, comorbidity, vital signs, laboratory test results, scoring systems, and treatment information on the first day of ICU admission. Four-year mortality was set as the study outcome. We used the ML methods of logistic regression (LR), artificial neural network (NNET), naïve bayes (NB), gradient boosting machine (GBM), adapting boosting (Ada), random forest (RF), bagged trees (BT), and eXtreme Gradient Boosting (XGB). The prognostic capacity and clinical utility of these ML models were compared using the area under the receiver operating characteristic curves (AUC), calibration curves, and decision curve analysis (DCA).ResultsOf 7,368 patients in MIMIC-III included in the final cohort, a total of 1,337 (18.15%) patients died during a 4-year follow-up. Among 65 variables extracted from the database, a total of 25 predictors were selected using recursive feature elimination and included in the subsequent analysis. The Ada model performed best among eight models in both discriminatory ability with the highest AUC of 0.801 and goodness of fit (visualized by calibration curve). Moreover, the DCA shows that the net benefit of the RF, Ada, and BT models surpassed that of other ML models for almost all threshold probability values. Additionally, through the Ada technique, we determined that red blood cell distribution width (RDW), blood urea nitrogen (BUN), SAPS II, anion gap (AG), age, urine output, chloride, creatinine, congestive heart failure, and SOFA were the Top 10 predictors in the feature importance rankings.ConclusionsThe Ada model performs best in predicting 4-year mortality after cardiac surgery among the eight ML models, which might have significant application in the development of early warning systems for patients following operations.

Artificial intelligence for clinical decision support for monitoring patients in cardiovascular ICUs: A systematic review

“…Areas where RL has been applied, that are relevant for cardiovascular monitoring include targeting of measurements during monitoring and choosing, timing and dosing of treatment steps. Many diagnostic and prognostic tasks in the healthcare domain are facilitated through the use of a variety of supervised ML models including logistic regression (LR), support vector machines (SVM), and ensemble methods such as random forest (RAF) and extra trees (39)(40)(41)(42). This group of AI algorithms are often applied on time-independent tabular patient information.…”

Machine learning‐based prediction of 1‐year mortality in hypertensive patients undergoing coronary revascularization surgery

“…Like our study, a recent survey reported the 25 important predictors selected by the RF algorithm for mortality after cardiac surgery, which include chronic HF, mechanical ventilation, sodium, blood pressure, Hb, age, creatinine, renal failure, dyslipidemia, and glucose. 10 In another study, XGB selected the serum creatinine, weight, age, and EF as the most important predictor for in-hospital and 30-day mortality of cardiac surgery. 69 Besides, age, renal disease, chronic heart failure, and hyperlipidemia were selected as influential factors of long-term survival in elderly patients with CABG by various ML algorithms.…”

“…In this study, LR was found to be the second best-performing ML model with a slight difference in accuracy from the Ada (AUC = 0.797). 10 As LR is sometimes considered a traditional model and not an ML algorithm, a meta-analysis concluded that ML was superior to the LR model in terms of mortality prediction after cardiac surgery, with a nonsignificant trend toward the better predictive ability of each ML algorithm. Nevertheless, the clinical importance of such an enhancement remains challenging.…”

“…9 Besides traditional risk scores, machine learning (ML)developed models are getting attention for outcome prediction after cardiac surgeries. 10 However, there are controversies about the accuracy of ML models compared to risk scores currently being used. 11 Knowing the greater need for mortality prediction in the hypertensive population, we aimed to use and compare different ML methods to predict 1-year mortality of hypertensive patients after isolated CABG.…”

Predicting the unpredictable in cardiothoracic surgery

“…The efficacy of ML models, supplementing various societal risk models, for predicting outcomes in other types of surgeries like aortic valve replacement has been demonstrated by Kilic et al [5]. In fact, ML models have even been used for predicting long-term mortality following cardiac surgery [6].…”