Dataset Information

Utilizing Machine Learning Methods for Preoperative Prediction of Postsurgical Mortality and Intensive Care Unit Admission.

ABSTRACT: OBJECTIVE:To compare the performance of machine learning models against the traditionally derived Combined Assessment of Risk Encountered in Surgery (CARES) model and the American Society of Anaesthesiologists-Physical Status (ASA-PS) in the prediction of 30-day postsurgical mortality and need for intensive care unit (ICU) stay >24?hours. BACKGROUND:Prediction of surgical risk preoperatively is important for clinical shared decision-making and planning of health resources such as ICU beds. The current growth of electronic medical records coupled with machine learning presents an opportunity to improve the performance of established risk models. METHODS:All patients aged 18 years and above who underwent noncardiac and nonneurological surgery at Singapore General Hospital (SGH) between 1 January 2012 and 31 October 2016 were included. Patient demographics, comorbidities, preoperative laboratory results, and surgery details were obtained from their electronic medical records. Seventy percent of the observations were randomly selected for training, leaving 30% for testing. Baseline models were CARES and ASA-PS. Candidate models were trained using random forest, adaptive boosting, gradient boosting, and support vector machine. Models were evaluated on area under the receiver operating characteristic curve (AUROC) and area under the precision-recall curve (AUPRC). RESULTS:A total of 90,785 patients were included, of whom 539 (0.6%) died within 30 days and 1264 (1.4%) required ICU admission >24?hours postoperatively. Baseline models achieved high AUROCs despite poor sensitivities by predicting all negative in a predominantly negative dataset. Gradient boosting was the best performing model with AUPRCs of 0.23 and 0.38 for mortality and ICU admission outcomes respectively. CONCLUSIONS:Machine learning can be used to improve surgical risk prediction compared to traditional risk calculators. AUPRC should be used to evaluate model predictive performance instead of AUROC when the dataset is imbalanced.

SUBMITTER: Chiew CJ

PROVIDER: S-EPMC7668340 | biostudies-literature | 2020 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Utilizing Machine Learning Methods for Preoperative Prediction of Postsurgical Mortality and Intensive Care Unit Admission.

Chiew Calvin J CJ Liu Nan N Wong Ting Hway TH Sim Yilin E YE Abdullah Hairil R HR

Annals of surgery 20201201 6

<h4>Objective</h4>To compare the performance of machine learning models against the traditionally derived Combined Assessment of Risk Encountered in Surgery (CARES) model and the American Society of Anaesthesiologists-Physical Status (ASA-PS) in the prediction of 30-day postsurgical mortality and need for intensive care unit (ICU) stay >24 hours.<h4>Background</h4>Prediction of surgical risk preoperatively is important for clinical shared decision-making and planning of health resources such as ...[more]

PMID: 30973386

Similar Datasets

Project description:BackgroundTo develop a highly discriminative machine learning model for the prediction of intensive care unit admission (>24h) using the easily available preoperative information from electronic health records. An accurate prediction model for ICU admission after surgery is of great importance for surgical risk assessment and appropriate utilization of ICU resources.MethodData were collected retrospectively from a large hospital, comprising 135,442 adult patients who underwent surgery except for cardiac surgery between 1 January 2014, and 31 July 2018 in China. Multiple existing predictive machine learning algorithms were explored to construct the prediction model, including logistic regression, random forest, adaptive boosting, and gradient boosting machine. Four secondary analyses were conducted to improve the interpretability of the results.ResultsA total of 2702 (2.0%) patients were admitted to the intensive care unit postoperatively. The gradient boosting machine model attained the highest area under the receiver operating characteristic curve of 0.90. The machine learning models predicted intensive care unit admission better than the American Society of Anesthesiologists Physical Status (area under the receiver operating characteristic curve: 0.68). The gradient boosting machine recognized several features as highly significant predictors for postoperatively intensive care unit admission. By applying subgroup analysis and secondary analysis, we found that patients with operations on the digestive, respiratory, and vascular systems had higher probabilities for intensive care unit admission.ConclusionCompared with conventional American Society of Anesthesiologists Physical Status and logistic regression model, the gradient boosting machine could improve the performance in the prediction of intensive care unit admission. Machine learning models could be used to improve the discrimination and identify the need for intensive care unit admission after surgery in elective noncardiac surgical patients, which could help manage the surgical risk.

Project description:IntroductionAccurate surgical risk prediction is paramount in clinical shared decision making. Existing risk calculators have limited value in local practice due to lack of validation, complexities and inclusion of non-routine variables.ObjectiveWe aim to develop a simple, locally derived and validated surgical risk calculator predicting 30-day postsurgical mortality and need for intensive care unit (ICU) stay (>24?hours) based on routinely collected preoperative variables. We postulate that accuracy of a clinical history-based scoring tool could be improved by including readily available investigations, such as haemoglobin level and red cell distribution width.MethodologyElectronic medical records of 90?785 patients, who underwent non-cardiac and non-neuro surgery between 1 January 2012 and 31 October 2016 in Singapore General Hospital, were retrospectively analysed. Patient demographics, comorbidities, laboratory results, surgical priority and surgical risk were collected. Outcome measures were death within 30 days after surgery and ICU admission. After excluding patients with missing data, the final data set consisted of 79?914 cases, which was divided randomly into derivation (70%) and validation cohort (30%). Multivariable logistic regression analysis was used to construct a single model predicting both outcomes using Odds Ratio (OR) of the risk variables. The ORs were then assigned ranks, which were subsequently used to construct the calculator.ResultsObserved mortality was 0.6%. The Combined Assessment of Risk Encountered in Surgery (CARES) surgical risk calculator, consisting of nine variables, was constructed. The area under the receiver operating curve (AUROC) in the derivation and validation cohorts for mortality were 0.934 (0.917-0.950) and 0.934 (0.912-0.956), respectively, while the AUROC for ICU admission was 0.863 (0.848-0.878) and 0.837 (0.808-0.868), respectively. CARES also performed better than the American Society of Anaesthesiologists-Physical Status classification in terms of AUROC comparison.ConclusionThe development of the CARES surgical risk calculator allows for a simplified yet accurate prediction of both postoperative mortality and need for ICU admission after surgery.

Project description:BackgroundsThe in-hospital mortality in lung cancer patients admitted to intensive care unit (ICU) is extremely high. This study intended to adopt machine learning algorithm models to predict in-hospital mortality of critically ill lung cancer for providing relative information in clinical decision-making.MethodsData were extracted from the Medical Information Mart for Intensive Care-IV (MIMIC-IV) for a training cohort and data extracted from the Medical Information Mart for eICU Collaborative Research Database (eICU-CRD) database for a validation cohort. Logistic regression, random forest, decision tree, light gradient boosting machine (LightGBM), eXtreme gradient boosting (XGBoost), and an ensemble (random forest+LightGBM+XGBoost) model were used for prediction of in-hospital mortality and important feature extraction. The AUC (area under receiver operating curve), accuracy, F1 score and recall were used to evaluate the predictive performance of each model. Shapley Additive exPlanations (SHAP) values were calculated to evaluate feature importance of each feature.ResultsOverall, there were 653 (24.8%) in-hospital mortality in the training cohort, and 523 (21.7%) in-hospital mortality in the validation cohort. Among the six machine learning models, the ensemble model achieved the best performance. The top 5 most influential features were the sequential organ failure assessment (SOFA) score, albumin, the oxford acute severity of illness score (OASIS) score, anion gap and bilirubin in random forest and XGBoost model. The SHAP summary plot was used to illustrate the positive or negative effects of the top 15 features attributed to the XGBoost model.ConclusionThe ensemble model performed best and might be applied to forecast in-hospital mortality of critically ill lung cancer patients, and the SOFA score was the most important feature in all models. These results might offer valuable and significant reference for ICU clinicians' decision-making in advance.

Dataset Information

Utilizing Machine Learning Methods for Preoperative Prediction of Postsurgical Mortality and Intensive Care Unit Admission.

Publications

Utilizing Machine Learning Methods for Preoperative Prediction of Postsurgical Mortality and Intensive Care Unit Admission.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets