Search

ObjectivesMachine learning (ML) classification tools are known to accurately predict many cardiac surgical outcomes. A novel approach, ML-based survival analysis, remains unstudied for predicting mortality after cardiac surgery. We aimed to benchmark performance, as measured by the concordance index (C-index), of tree-based survival models against Cox proportional hazards (CPH) modeling and explore risk factors using the best-performing model.Methods144,536 patients with 147,301 surgery events from the Australian and New Zealand Society of Cardiac and Thoracic Surgeons (ANZSCTS) national database were used to train and validate models. Univariate analysis was performed using Student's T-test for continuous variables, Chi-squared test for categorical variables, and stratified Kaplan-Meier estimation of the survival function. Three ML models were tested, a decision tree (DT), random forest (RF), and gradient boosting machine (GBM). Hyperparameter tuning was performed using a Bayesian search strategy. Performance was assessed using 2-fold cross-validation repeated 5 times.ResultsThe highest performing model was the GBM with a C-index of 0.803 (0.002), followed by RF with 0.791 (0.003), DT with 0.729 (0.014), and finally CPH with 0.596 (0.042). The 5 most predictive features were age, type of procedure, length of hospital stay, drain output in the first 4 h (ml), and inotrope use greater than 4 h postoperatively.ConclusionTree-based learning for survival analysis is a non-parametric and performant alternative to CPH modeling. GBMs offer interpretable modeling of non-linear relationships, promising to expose the most relevant risk factors and uncover new questions to guide future research.

Abstract Background Hyperbilirubinemia following cardiac surgery is a common phenomenon and is of emerging interest in prognostic factor research. This systematic review and meta-analysis evaluated the association between post-operative hyperbilirubinemia (PH) and mortality and morbidity in cardiac surgery patients. Methods Ovid Medline and Ovid Embase were searched from inception to July 2020 for studies evaluating the prognostic significance of PH following cardiac surgery. Maximally adjusted odds ratios (OR) with associated confidence intervals were obtained from each study and pooled using random effects inverse variance modelling to assess in-hospital mortality. Standardised mean differences were pooled to assess Intensive Care Unit (ICU) and hospital length of stay (LOS). Qualitative analysis was performed to assess ventilation requirements and long-term mortality. Meta-regression was used to assess inter- and intra-study heterogeneity. Results 3251 studies satisfied the selection criteria, from which 12 studies incorporating 3876 participants were included. PH significantly predicted in-hospital mortality with a pooled OR of 7.29 (95% CI 3.53, 15.09). Multiple pre-defined covariates contributed to the prognostic significance of PH, however only aortic cross-clamp time (p

Machine learning (ML) is increasingly applied to predict adverse postoperative outcomes in cardiac surgery. Commonly used ML models fail to translate to clinical practice due to absent model explainability, limited uncertainty quantification, and no flexibility to missing data. We aimed to develop and benchmark a novel ML approach, the uncertainty-aware attention network (UAN), to overcome these common limitations. Two Bayesian uncertainty quantification methods were tested, generalized variational inference (GVI) or a posterior network (PN). The UAN models were compared with an ensemble of XGBoost models and a Bayesian logistic regression model (LR) with imputation. The derivation datasets consisted of 153,932 surgery events from the Australian and New Zealand Society of Cardiac and Thoracic Surgeons (ANZSCTS) Cardiac Surgery Database. An external validation consisted of 7343 surgery events which were extracted from the Medical Information Mart for Intensive Care (MIMIC) III critical care dataset. The highest performing model on the external validation dataset was a UAN-GVI with an area under the receiver operating characteristic curve (AUC) of 0.78 (0.01). Model performance improved on high confidence samples with an AUC of 0.81 (0.01). Confidence calibration for aleatoric uncertainty was excellent for all models. Calibration for epistemic uncertainty was more variable, with an ensemble of XGBoost models performing the best with an AUC of 0.84 (0.08). Epistemic uncertainty was improved using the PN approach, compared to GVI. UAN is able to use an interpretable and flexible deep learning approach to provide estimates of model uncertainty alongside state-of-the-art predictions. The model has been made freely available as an easy-to-use web application demonstrating that by designing uncertainty-aware models with innately explainable predictions deep learning may become more suitable for routine clinical use.

Background. Cardiac troponin is a highly specific and widely available marker of myocardial injury, and elevations in cardiac transplant donors may influence donor selection. We aimed to investigate whether elevated donor troponin has a role as a prognostic biomarker in cardiac transplantation. Methods. In a systematic review and meta-analysis, we searched MEDLINE, Embase, and the Cochrane Library, without language restriction, from inception to December 2020. We included studies reporting the association of elevated donor troponin with recipient outcome after cardiac transplant. We generated summary odds ratios and hazard ratios for the association of elevated donor troponin with short- and long-term adverse outcomes. Methodological quality was monitored using the Quality In Prognosis Studies tool, and interstudy heterogeneity was assessed using a series of sensitivity and subgroup analyses. Results. We included 17 studies involving 15 443 patients undergoing cardiac transplantation. Elevated donor troponin was associated with increased odds of graft rejection at 1 y (odds ratio, 2.54; 95% confidence interval, 1.22-5.28). No significant prognostic relationship was found between donor troponin and primary graft failure, short- to long-term mortality, cardiac allograft vasculopathy, and pediatric graft loss. Conclusions. Elevated donor troponin is not associated with an increased short- or long-term mortality postcardiac transplant despite increasing the risk of graft rejection at 1 y. Accordingly, an elevated donor troponin in isolation should not exclude donation.

Objective(s)To use machine learning (ML) to predict short-term requirements for invasive ventilation in patients with COVID-19 admitted to Australian intensive care units (ICUs).DesignA machine learning study within a national ICU COVID-19 registry in Australia.ParticipantsAdult patients who were spontaneously breathing and admitted to participating ICUs with laboratory-confirmed COVID-19 from 20 February 2020 to 7 March 2021. Patients intubated on day one of their ICU admission were excluded.Main outcome measuresSix machine learning models predicted the requirement for invasive ventilation by day three of ICU admission from variables recorded on the first calendar day of ICU admission; (1) random forest classifier (RF), (2) decision tree classifier (DT), (3) logistic regression (LR), (4) K neighbours classifier (KNN), (5) support vector machine (SVM), and (6) gradient boosted machine (GBM). Cross-validation was used to assess the area under the receiver operating characteristic curve (AUC), sensitivity, and specificity of machine learning models.Results300 ICU admissions collected from 53 ICUs across Australia were included. The median [IQR] age of patients was 59 [50-69] years, 109 (36%) were female and 60 (20%) required invasive ventilation on day two or three. Random forest and Gradient boosted machine were the best performing algorithms, achieving mean (SD) AUCs of 0.69 (0.06) and 0.68 (0.07), and mean sensitivities of 77 (19%) and 81 (17%), respectively.ConclusionMachine learning can be used to predict subsequent ventilation in patients with COVID-19 who were spontaneously breathing and admitted to Australian ICUs.